I continue to think the right way to do age verification is at the operating system level. I am happy to provide a digital ID to an operating system and happy to allow an API call to confirm my age. I am not ok with any kind of distributed storage of my identity or eliminating anonymity when accessing content deemed objectionable.

My identity is already put at risk in so many ways— just look at how thoroughly compromised social security numbers are. There are data breaches all the time. And this government seems to be tirelessly gating my rights behind my identification and making it harder and harder to get a valid government ID. So why would I want to offer my identity to all these websites and force them to store it?

And there’s absolutely no way I should have to have my real identity so easily identified with every single piece of media I interact with.

The details here matter, but the concept of having a form of age-gating for content seems ok to me.

The Texas law is absolute shit, no surprise. And SCOTUS is absolute shit, no surprise (well, 6/9 are). But age verification seems fine— we should do this right.

I really wish Apple and Google would just fucking step it up and make a privacy protecting mechanism for this now that we know it’ll be the law of the land in so many states soon. Point the way so that legislatures that want to gate this content know how to do so without compromising identity or security and without creating the need for 400 different ways to be compliant.

I think The Future Doesn’t Suck has it mostly right. While I think the current crop of AI has made it well passed “works good enough to actually bend the efficiency curve”, I don’t think the result of this will be reduced headcount. Like with most things, simple supply and demand models are still at work here. When it becomes less expensive to deliver something, demand goes up. The fact that I can write more code, whether it’s delivering features or improving bugs, means that the expectation from users of software products will be higher. They’ll expect more features, at higher quality, delivered faster than ever before. And they’ll want that because if I don’t offer it, someone else will.

While I’d like a four day work week, making a white collar worker more efficient at building zero marginal cost products is not going to get me there. Customers will choose software that has engineers who work five days a week but produce 6+ days worth of improvement over those that work 4 days and get 5 days of improvement.

This is not literally always true or endlessly true, but it’s true enough and the accessibility of AI is enough that in competitive markets I expect this to be the case.

I can do my job more efficiently and effectively, but so can everyone I’m competing against. So I’ll have to keep chasing them. Doing a good job still means delivering a better product, but the bar has been raised.

I don’t think I’m going to work less– I think I’m going to ship better products. What I’m actually hopeful for is that my job will be more fun. The jury is still out on that one.

The more general lesson I take from Pocket (as I have taken from so many such acquisitions-and-eventual-shutdown stories before it) is this: Not all businesses need to be large, with paths to continual growth. Quite the contrary. The expectations of scale, continual growth, or especially both together, become perverse and destructive when they become defaults and norms. Some few businesses will grow very large and will experience continued growth for long periods of time. Most, however, will not, and the attempt to do so will lead them away from serving their existing customers along the way, and just as importantly will usually fail.

— Chris Krycho, writing about the failure of Pocket.

I don’t think anyone is surprised by the idea that “not all businesses need to be large” (nor is Chris— I’m using his post and point as a jumping off point). I think the corrosive problem of the zero interest rate internet world was the destruction of all non-scale business models.

It’s still hard to build any form of consumer software and charge for it. And the only way to be a business that reaches an end to its growth is to make enough money all along the way. But we’ve taught consumers to expect to spend nothing, leaving only mass advertising or charging businesses as the only option to build a business. And, at least at in the past, the initial costs of building software are high enough that many companies have taken on forms of debt that cannot be satisfied with mere profitability.

The end of living on the lower bounds of interest rates and the reduction in costs associated with building software will hopefully change all of this, but I’m not holding my breath. Consumers have been taught software should be free. We’re bombarded with people online who think that the marginal costs for delivering goods and services should fully determine the price (see all leftist pharmaceutical discussions). I remember telling somone I paid $10 for a calendaring app (this is pre-SaaS/subscription mania) and they were shocked. When I replied, “I use that app dozens of times a day, every single day, and every time I use it I have a better experience than with the alternative,” it only kind of sort of made sense to them.

Everyone thinks everything should be free because Google and Facebook found their way into a magical business model with winner take-all dynamics. But we already have the winners, so we need to do something else now.

Because MarsEdit doesn’t exist on iOS and I’m traveling without my personal laptop, I find myself only posting photos to Instagram. Hopefully, I’ll care to post on my blog soon. The problem is the Micro.blog app absolutely destroys photos— resolution gets fucked and the color space gets fucked. It also makes it too hard to insert Markdown for photos versus HTML, and I rely on a Markdown render hook for lightboxes and galleries.

Micro.social is a bit better, but also lacks the quick inline Markdown to place photos in longer posts to make galleries work better and doesn’t have the nice AI alt text feature, which really reduces the barrier to posting.

Netflix has been absolutely horrible at managing renewals— and the same is true for quite a few other streamers. I have completely dropped out of multiple shows that I liked because I couldn’t remember them by the time they returned.

Yes, yes, COVID and the strikes. But the above article mentioned another factor I hadn’t considered— shorter seasons mean actors have to book more work which means more complex shooting schedules.

I was so happy to hear about how The Pitt was not going to fall into this trap. The Pitt’s show structure also meant that actors got paid, well, even those without major parts. Because the show took place in “real time”, with each hour of the show being an hour in the same day in the same location, even background extra-type actors were in many scenes over many episodes.

My conclusion is different than Mike Goldstein’s (or what he purports to be John Arnold’s)- I think that we basically are doing close to as best we can within the resourcing and talent available to operate schools. Without changing their resources far more dramatically than philanthropy or changing goals, this is about as good as it gets.

Resource equity remains the most important game in town because it’s the most important difference between successful and functioning schools and dysfunctional places. But dysfunction does not evaporate with resources— they need both resources and time, beyond our political patience horizon.

And lastly, schools can’t improve much if society drags. If the opportunities and supports cease outside of school walls, nothing will change.

Why are there so few professors that are Republicans? The GOP demonizes and rejects the conclusions of the scientific process– like climate change, vaccines, or tariffs. You can’t train for a decade in science and see yourself supported in a party that governs against reality.

It’s just like why Trump loses more in court– he does more illegal shit.

Right now, the GOP has decided that the correct response to new information that’s incongruous with their beliefs are to reject the information and not update their beliefs. And the track record for their beliefs aligning with knowledge is not great. For the GOP, reality is whatever they assert it is. This is fundamentally incompatible with modern traditions in both science and law.

I wouldn’t say that the Democrats are right, and certainly not right all the time. But they are significantly less hostile to the idea that we can, in fact, learn new information, and we can, in fact, make different choices as a result.

The mission of MAGA is to do to all our institutions what occurred in the American church. We are replacing doctors and public health experts with the equivalent of televangelist and megachurch prosperity gospel drifters. And we are doing the same with education, environmental regulation, communications, trade, securities, social security, internal revenue, etc. It’s what we’ve done to school board elections, city councils, and every part of local democracy as well.

In every corner of our government, we are replacing expertise, local knowledge, and dedicated public service with something that looks slick, compels the ignorant, and is designed to turn us from a model of participatory communities to mass culture consumption. It will enrich a few and impoverish us all– spiritually and economically.

Inspired by Ultranurd’s WWDC 2025 Wishlist, here’s mine. This is not a list focused on the Apple Developer community– it’s a list focused on what I, a user, wants out of macOS (mostly) software ¹.

1. Like the aforementioned wishlist, my number one ask is for macOS to get a strong, first party package manager, with none of the bullshit of the Mac App Store. Even Windows has one now, Homebrew is acceptable, but kind of a mess and a bit nuts to rely on for key platform level tools. If you care about developers as users of your platform, this is a key element. Swift has a package manager– it’s an obviously important feature of a healthy programming language. But package management is also a critical component of healthy systems.

2. Apple should bite the bullet and enter the search space. Their deal with Google is on life support, and AI-based search is an opportunity to reset the board. Apple has gone from a leader in some types of search (remember when Spotlight and Apple Mail search were huge advantages that Apple held versus hopeless horrible messes that seem worse than even naive Sqlite full text search?) to a laggard. They are seeking services revenue wherever it can be found. Search is an area to stretch their AI capabilities, their web services capabilities, their privacy focus, and their advertising revenue. Just do it– it’s time to play.

3. Fix Notifications– they’re a mess, but especially Notification Center on macOS. Notification state (read/presence whatever) needs to sync across devices. Notification state should be clearable by an app (for example, how often do I have a notification from Messages for a message I’ve already read in Messages? Always). Notifications should be easier to directly take action on.

4. Shared Albums in Apple Photos should be full quality photos. This is why I never use this feature. And speaking of Shared Albums…

5. Instagram the Shared Albums feed. How many people know about liking and commenting on Photos in Shared Albums? And that there’s an Activity Feed? It feels like Apple is so close to a strong, private, photo sharing network that’s just a little short, yet could be incredible for those of us who have lost what Meta’s products used to be– a decent place to share things with friends. Or buy Retro.app, which continues to be a favorite of mine but no one uses it.

6. Make ambitious updates of side project apps. Let’s see things like Clips, Invites, and the return (I’m dreaming here, right?!) of Music Memos. All of these apps should have 2-3 great application developers inside of Apple supporting them year round. They should be the first to be involved in any framework changes for iOS or new UI paradigms. They should release with best in case support for Apple Intents and Shortcuts and all of that. They should be giving feedback on things like how to migrate to Swift 6 with strict concurrency. They should be howling the documentation teams and helping to improve it. Keeping a slate of 5-10 great, purpose built apps on each platform that are not apart of the platform teams and can’t use private APIs is a huge opportunity to dog food. And these applications, even when they don’t set the world on fire, are often genuinely great. Even better, open source these applications. No more demo code– show us how things work for real, in context. Ok, this item maybe is for Apple developers.

7. Increase iCloud storage, and push iCloud features to the next stage. Will iCloud ever be as feature packed as other cloud storage providers? Probably not. But it should keep working its way closer and closer. This will also improve the APIs that Dropbox and others are being pushed to use.

8. Invest more in Focus Modes– how? I’m not sure. Why? Because this feature has been the greatest improvement in iOS since copy and paste.

9. Provide some hint toward a redesigned Apple Watch. I’m so happy with the data, which nothing else seems to offer. But I’m so tired of the design. I find myself resenting wearing the Watch while being unwilling to give it up. Just give us something fresh this year.

10. Add weightlifting / more specific exercise tracking to Apple Fitness. I want to write what I did at the gym, what weights I used, reps, sets, etc. I want Apple to try and do things like tell me “If you’re trying to build muscle, it looks like you need more reset between sets,” based on analyzing my heart rate or wrist temperature changes, or something with the Watch sensors.

11. Redesign the Home app again. What a shit show.

When the Insurrectionists showed up in full tactical gear and assault rifles, stormed the US Capitol, pommeled police officers, and destroyed government property, not only did Trump fail to act, he ultimately pardoned them en masse.

As people in flip flops peaceably surrounded ICE agents, who were the ones in tactical gear with guns, protesting their cruelty in using state power for violence and detainment, Trump is preparing to send a literal army with the intent to terrorize, kill, detain, and likely bring outrageous charges against the protestors.

Trump is a depraved mad king supported by the irredeemable. They lack a critical element of humanity, and are driven by an unjustified hatred and grievance born of ignorance and manipulation.

These are the Ugly Americans, dead set on destroying our country. They want to terrorize a free people and destroy our institutions and our values. Why? Because that’s what their godless churches told them. Because that’s what incredible rich and powerful people paid people to tell them. Because the world keeps changing like it always did and that frightens them. Because they lack the ability to love anyone as much as they, deep down, loathe themselves. Because TV entertainment matters more than purpose or morality. Because they are the kind of people who only try to win whatever game is in front of them rather than question why they’re playing or who made the rules.

Because these are the people who think they “dare to disturb the universe” by enabling a gang.

These are the Ugly Americans, who want other people’s joy to die, especially if it looks different than their own hopes and dreams. These are the Ugly Americans, who would rather be sad and dominate you than join you as an equal in prosperity.

These are the worst people, who enable evil. They are evil. We live in a place filled with evil.

Fuck Trump.

This Memorial Day involved a bunch of random activities. Including:

Setting up a new electronics work station

Who knows if this hobby will stick? But I’ve decided it’s time to get to work on a breadboard and with a soldering iron. I’m hoping to build some guitar pedals as a starting point. Send me your fun electronics project ideas!

Power Washing

I haven’t properly cleaned this deck since we moved in almost 8 years ago. It’s way past time. I spent over an hour, and there’s still so much more to do. Also, there are so many wasps out there.

Ribs

When Whole Foods has ribs on sale for Memorial Day weekend, you make some ribs.

Misc.

We watched the first two episodes of The Last of Us Season 2. It’s… fine.

I played a lot of guitar.

We looked at a lot of dogs to adopt.

We’ve had to say no to the first pup, even though we were approved for adoption. They slipped in at the last moment that she was heartworm positive and they had not used the recommended treatment course. This would have meant thousands of dollars, but more importantly, three months of crate rest.

I just didn’t feel comfortable bringing a new dog into my home and spending the first three months of our relationship constantly begging a puppy to be calm and stay in their crate so their heart rate doesn’t get up and they get an embolism.

Potty training under these conditions, and building a healthy relationship with my dog, felt like the wrong choice for us.

Hoping we find another pup soon.

Manuel is great, but his take on the web sometimes is really frustrating.

The web, the “information system that enables content sharing over the Internet through user-friendly ways meant to appeal to users beyond IT specialists and hobbyists”, the one that was here from the beginning, before Google, before ad tech, before socials, that web is here to stay.

Yeah, I was here for that web. Before Google and ad-tech there were still ads, but more importantly, the web was quiet and discovery was hard. The way we found places to spend time online were still via search and social sharing. Search existed well before Google, and before search, we used directories. And before social media platforms, we still discovered web content by clicking links in other web content.

The original web hosting and blogging platforms that were most successful all were social networks. LiveJournal, Blogspot, heck even Geocities to an extent all had forms of native sharing of links to other sites made on their services. They all succeeded not just on making it easier to host a website, but through providing a network effect. You got the most out of LiveJournal by following other LiveJournal accounts. You chose LiveJournal or Xanga based on where your friends and people were.

Before (and concurrently, of course) personal blogging, so much of the best “web” content was found on topical forums, and before (and concurrently, of course) that it was newsgroups. The best of the web is written by people for other people where other people share and see it. The “web” part is links, that made all of this content accessible and shareable and connected anywhere you went.

Search and social sharing have always been how we discover content on the web. There is no web without both. There are websites, but the actual “network” is built on links, and we find links when people share those links or use some other aggregator to find links they’re interested in.

The idea that search is not important to personal websites is crazy talk– it was and is still a great way to find webpages. The idea that social sharing isn’t important to websites is crazy talk– I love to find and follow links from blog posts, but that’s no more or less social than links posted wherever people who share my interests post links.

I agree with so much of what Manuel writes and respect him a ton, but he also takes a stand on the internet that based in a countercultural ideal that bleeds into an ahistorical picture of an early internet that never existed and would have never existed because it doesn’t work. We can reject how todays social media platforms make us feel, as products or companies, without forgetting that building a network where people can discover both content and people relevant to their interests is the whole of the web we care about. Corporate, small, quiet, indie, fedi-, whatever– all forms of a web that works is a web that connects and makes forming connections easier.

Links are what make it a web and not a document. Links are the key to hypermedia of all kinds. Any attack on links is an attack on the web.

Importantly, search is one of the most powerful ways to find the right links. And in a rare virtuous cycle, since the dawn of Google, links are currency. Not only does search help you gather links, providing links improves search. Search has incentivized connecting the web. The more connected content is, the more likely it is to appear in search. In turn, search is improved, becoming more likely to surface relevant content on the whole.

It is a huge mistake to write off the importance of search as though it is only meaningful for content farming or commercial use of publishing on the internet.

Manuel writes:

Because the true soul of the web is not the never-ending quest to make money, but rather the desire to create, to express ourselves, to connect.

I just think Manuel is completely wrong about how the connect part happens.

So here’s what I learned today:

• I still love pretty much every American G&L guitar I play. I think I just really get along with their fretwork.
• Nothing quite gets that Fender strat sound like a Fender strat. It’s just not the sound I need right now.
• I like the Benson Vinny almost as much as I liked the Monarch (may have been a difference in room), but boy these are pricey. And I’m not sure how it’ll work for the higher gain stuff I pay. Still, great amp, and great to play three different guitars into an amp I basically knew already in order to compare them. Ultimately, no new guitar today.

I really liked one G&L Legacy’s sound ¹, but I didn’t like how it looked. It also needed to be rewired (neck pickup was dead, possibly middle as well– it had a weird tilt-blade switch that I think was splitting the bridge coil, but that meant positions 3, 4, and 5 worked, variably based on the tilt).

The other G&L Legacy ² I played looked great, but someone replaced the humbucker with some kind of Seymour Duncan that I just sounded really bad. There was a lot of fizz and flub and lack of clarity. Maybe new pickups would have solved the problem, but I don’t love taking that gamble.

The last guitar I spent a good chunk of time with was a limited Fender with some kind of roasted maple neck I think. It looked quite dark, like mahogany, but because there was no separate fingerboard I’m guessing it was “roasted maple”, which I’ve never played. {Update: The neck was a single, solid piece of rosewood (including the headstock). I quite liked this!} It sounded great, and it played great even though it had medium/vintage frets, which I’m learning I generally don’t like. But it sounded super strat like. It just was full of that vintage, single coil, quack. It was exactly great at what it does, but that’s not the sound I need to add to my stable today.

This was my first time at Atomic Music. They had a collection of amps I would have died for in high school. And while they have a large, packed in guitar stock, it was not as fun for me. There were just maybe 5-6 guitars of the hundreds that spoke to me. The organization was pretty chaotic, which meant it was kind of hard to find what I was looking for anyway. It was far too busy to look at pedals, and the pedal stock was kind of a mess so it would have been pretty disruptive on a busy Saturday to try stuff out.

I was glad I went, and, especially if I was amp shopping, I’d go back again. I need to find a G&L dealer with extensive stock. Maybe I’ll go to Guitar Center tonight– surprisingly, the last time I went they had a lot of guitars that struck my fancy.

Note Wiring is not really complete– I need some additional organization and cable lengths to make things clean and ideal, but I think this is the core layout.

Signal chain

Polytune 3 Mini -> Keeley OctaPsi -> BSRI Magawa -> (Amp) -> Walrus Julia -> Chase Bliss Billy Strings Wombtone -> Walrus ARP-87 -> Walrus Fundamental Ambient -> (Amp)

(Amp) is mostly the Dr. Z Maz 18 I have a UA Lion ‘68 (pictured) that I’m using sometimes for home recording and also intend to bring to gigs to enable an amp-less setup/backup if needed. I’m also not 100% sure I’ll always run the modulation/time stuff in the effects loop. In fact, I’ve mostly been running entirely into the front of the amp at rehearsals because it’s less hassle. But I think that I prefer this stuff in the loop, so I’m going to be less lazy about using a four cable method going forward.

The Dr. Z amp pedal is not a switcher– it’s a built in boost. To my ears, I don’t get much volume, likely because I have enough crunch set on the amp that I just get more compression, sustain, and gain when I slap that on.

The Walrus Canvas Power 15 is new, as is the pedalboard (Pedaltrain Classic 2, versus the original Pedaltrain I had– the tilt is really nice). Both are great so far.

Why are things arranged this way?

The spatial arrangement of the pedals is intended to make it easy to switch between the four-cable method, getting the dirt before the amp or the UA Lion, then getting the send from the effects loop into the Julia or going from the Lion to the Julia.

I put dirt on the bottom because that’s what I hit the most often. Magawa alone, amp boost alone, or both combined are my most common switches. That’s definitely followed by the octave on, which I use in a few songs to fatten a lead or a part heavy in diads. Typically, I am only using one modulation or time-based effect on at a time, so having them be a bit more distant is no problem. The Fundamental Ambient gets turned on the most, so it’s also in an easy spot at the edge.

What didn’t make the board

I continue to futz with the Slö and Fathom, the other two Walrus reverbs. I owned the Fundamental Ambient first and love everything about it. I’ve never been quite able to worm to the “big brothers” the Slö or Fathom. They may just not be for me. I honestly want to read more about both and look at other people’s settings and see if I can find a way to love them. If not, I’ll probably sell or trade them soon. I think I may want another reverb still– perhaps something more classic, perhaps something built into a delay like the Wampler Ethereal Delay and Reverb I traded (which I may get again, or perhaps try and EarthQuaker Devices Dispatch Master).

The Nano POG got pulled off when I got the OctaPsi, but it’s so damn good at what it does I don’t know if I’m willing to get rid of it yet.

Other uncertainties

I’m not so sure about the ARP-87. I think I liked the Wampler Ethereal Delay more, so like I said in Remainders about a reverb, I may be looking for a different delay, perhaps one with a reverb as well. I’m considering something like the JHS Flight Delay, too. I love the way the Julia sounds in general, but I’m still dialing it to find a setting that really fits with what I play and makes sense for a song. I suspect if I am ever going to get along with chorus, this will be the one, but I’m not quite there yet with it.

I got the Wombtone in my Mystery Box and immediately found a great setting for what I like a phaser to do and what I use it for in the band. Is it markedly better than the BSRI Resurrect Dead on Planet Jupiter– not for me, I use it in an almost identical fashion and don’t find it that much more meaningful. That said, it can do more and it’s a smaller enclosure, so when by chance I saw someone looking for the BSRI on Reddit I made the trade. I don’t need two phasers around, and BSRI will be at the Baltimore Pedal Show next month so I can always pick up another if I want and support a local builder.

Speaking of BSRI and reverbs, I’m pretty sure I want to look at the Soft Rains and the Smell of the Ground. I loved this pedal last year, and only didn’t get it because I was practically willing to buy everything BSRI had to show. Given my current reverb situation, I think it’s pretty likely I buy one of these next month.

What else am I considering? I think I want some more dirt options. The Magawa sounds and works great. The OctaPsi is untameable, like any Muff style pedal should be. But I still have interest in adding more textures and subtly. I think I’m most interested in a Browne Amplification Protein, the Boss OD-3, the Boss BD-2 or maybe instead a Boss Angry Driver JB-2, or possibly a Wampler Tumnus. The Benson Deep Sea Diver is also really interesting for another version of just “blow things up” in one pedal. I have tried hard not to gas over the Walrus Qi Etherealizer, because as much as I absolute adore Yvette Young, I can’t play even a little bit like her so I can’t “trust” the demos of what that pedal would do in my hands. But that being said, some of the sounds are absolutely crazy. I really like the idea of being able to create spacey pads of sound on guitar that are not just walls of fuzz. I also keep eyeing the EarthQuaker Devices Swiss Things. I do a fair amount of tap dancing sometimes to go from clean + Fundamental Ambient -> Magawa and amp boost on. It’d be kind of neat to have dirt in a loop so that I could set up any of my stacking how I need and one switch get to where I’m going. Plus, with the UA Lion, my Dr. Z amp, and the Valvetech amp, A/B/Y starts to get pretty interesting in terms of options.

Reddit recommended this post to me, with the following image:

As a reminder, this is what’s going on in my office, at the rehearsal studio, and who I am:

Ouch.

From ¹ Om Malik, quoting A Beginner's Guide to Japan :

Nowhere else I’ve been, in fact, are individuals so disengaged from the political domain; my Japanese friends assume they can no more address their leaders than they can a group of look-alike men in suits in a corporate boardroom with the doors locked and the curtains drawn. So they turn their backs on the public sphere, and make fantastic worlds out of their passions, counter-societies out of their hobbies.

This resonates. I think there’s a degree to which I have been turning away from the public sphere and towards my hobbies more and more over time. Some of this is changing where I am in my career. Some of this was moving away from Providence and consciously deciding to be uninvolved in local politics and news. Some of this was the despair of the first, and now second Trump presidency. But overtime, I’ve tried to put more time into things like playing with my band (just over a year now!), playing volleyball (just over 3 years now, with a 7 month break for injuries), my website, etc.

At times, I’ve worried this is unhealthy or immoral. Correction, no past tense– I worry this is unhealthy or immoral.

Other times I wonder if our greatest failures are from people who have turned entirely to the public sphere at the cost of their hobbies. How many bedroom warriors on Twitter, or living room screechers watching Fox News or OAN, or gym rat ragers listening to Ben Shapiro and Joe Rogan are unnaturally focused on the public sphere? How many of these people are so preoccupied with politics, and specifically what the government is or isn’t doing, that they’re failing to live their lives?

This form of bizarre populist politics as a hobby is toxic. As the kids these day say, go touch grass. They are so focused on oppressing people whose lives are not actually impacting them at all. I can’t believe that if they turned toward a different set of passions or hobbies and not the “public sphere”, all the manufactured outrage would disappear and the world would be better.

One of the primary critiques, and perhaps the only one that held much water over time, of Biden’s withdrawal from Afghanistan was that it left behind too many Afghanis who were in danger for being our allies. So it makes sense that Trump would remove the protective status from those who made it to the US.

It seems my hypocrisy knows no bounds.

There was a local guitar shop I sold gear to and had repairs done at. They had my email. I regularly checked their website to see what gear was available. I was considering signing up for lessons. I noticed starting in January that they listed way less gear on their site.

Turns out they closed lessons and the retail shop in March and moved to a repair only space. They only wrote about this on Instagram and Facebook, and made no note of this on their own website. They had closing sales, but never contacted people whose email they had from past used gear sales or repairs or announced this on their own domain.

I’m sad to lose a shop (it was just ok… but had promise), but more angry that businesses would communicate purely through Meta properties.

I have spent a fair amount of time “on the side” thinking about authorization rules over the last year. At work, things have gotten sufficiently complex that we are really straining our existing system. At least some of that comes from our choices, a lot of that comes from the domain complexity we have, and some of that comes from Phoenix only having an opinion in the current 1.8 release RC about “how this should be done” leaving us to make our own bad choices.

I’ve largely been insisting on maintaining an RBAC-driven or RBAC-like system. One thing I’ve learned in past versions of our applications is school districts want very fine-grained controls, almost on the level of individuals, and then quickly find out that maintaining permissions for individuals is just an untenable nightmare. But one thing that a school district does and can consistently know is what someone’s job title/role is.

What we’ve come up with are users have roles in an organization ¹. A user can have only one role in an organization, but they can have many of these… role-organization combinations. For example, I can be the principal supervisor of ABC Elementary and XYZ Elementary (two role-orgs) and the department head of Elementary Schools (a third role-org). Intuitively, I think this makes sense and is fairly easy to maintain.

But I have really struggled with language around this stuff. What are these role-organization combinations? Is organization or budget or any of these domain-ish terms correct, or is there a better technical term? Scopes come up often, but we extensively used that term elsewhere– now Phoenix supports scopes directly, so oops– and I found the term scope as having specific, inconsistent, and different definitions in different contexts. One doesn’t really have a role in a scope, but rather, has certain scopes by having a role.

Anyway, all this is preamble to say that while I haven’t read it yet, I’m glad that Tom MacWright linked to Zanzibar, something I completely missed when it came out in 2019. Like Tom, the distributed nature of Zanzibar is not of particular interest, but I’m hopeful that the features and concepts introduced will provide me with better language when talking about authorization going forward.

I think the solution to tech malaise, fog, or disappointment is to stop using your machines for casual distraction and go make something.

I’ve been working on demo tracks for my band since January, so I’m in new-to-me Ableton 12 every week. And there’s no way I could feel negative about technology when I can be this creative and have this much power.

I didn’t have to learn much to get started, and I’m learning constantly and it’s all easy and fun.

Writing software is fun. Writing blog posts are fun. Messing with my personal webpage is fun.

We don’t actually need a maglev train between DC and Baltimore, and we don’t even really need one along the NEC. It’s really a distraction from the HSR we should pursue. But:

1. Neither are going to happen in my lifetime, especially because of this administration.
2. This NIMBYism makes me want to support maglev.

Here are some of my current rules for email that help keep things tidy server side with Fastmail.

Newsletters

I have a few set of rules meant to capture newsletters. The first:

• If all of the following conditions are true:

  • A header called list-id exists.
  • Anywhere unsubscribe
  • Sender is not a contact

• Then

  • Move to Newsletters

• If all of the following conditions are true:

  • A header called List-Unsubscribe

• Then

  • Move to Newsletters

I then have the Newsletters folder delete all email greater than 31 days old.

Calendar Invites

Because calendar invites show up in my calendar app/apps of choice, having the email is really unhelpful. So:

• If all of the following conditions are true:
  • Anywhere filetype:calendar
• Then
  • Mark read
  • Move to Calendar Invites

I then have the Calendar Invites folder delete all email greater than 31 days old.

Political Email to Trash

I donate to some political causes, but their email is inescapable. Here are some conditions that have helped (note: this works for US Democratic candidates largely. We don’t talk about the other side). The first three grab common software used by Democrats, the next two are just examples of things I add when specific candidates tend to find their way around this:

• If any of the following conditions are true
  • A header called List-Unsubscribe contains actionkit
  • A header called List-Unsubscribe contains dccc
  • A header called List-Unsubscribe contains ngpvan
  • A header called Reply-to contains info@contact.kamalaharris.com
  • A header called Reply-to contains info@contact.joebiden.com

I edited this section to better match how I wrote the other sections, which was to mirror the GUI versus the actual search terms used to make these rules.

Specific Annoying Emails

Sometimes there’s a sender I want to subscribe to that has some specific spammy tendencies. For example, I don’t need my bank to tell me each time it initiates some scheduled transfers I have. I typically find the subject line of those common emails, mark as read and move to trash because I don’t need to get your “We started a Surprise Savings transfer” email four times a month. Similarly, notifications from online shopping I use commonly will get a header-based rule that moves them to Newsletters (I used to have a separate shopping folder, but it’s not worth it).

Sometimes, there’s a pernicious bit of spam. For example, I frequently got emails from a fake account that listed as payments <payments@bills.com> for these weird invoice things. No matter how many times I marked as spam, this particular joy kept getting through. So now it’s a rule that moves that email to spam.

Takeaway

There’s a whole lot of content in email headers (like those List-Unsubscribe fields) that are incredibly useful for automatically sorting your email and making things go away you don’t want to see. I don’t want to unsubscribe to every list I’m on, but I also don’t want my Inbox to be clogged up by non-urgent messages. I have very few folders and generally go Inbox -> Archive and use searching to find my mail. So by organizing a few top level folders for “Email I still basically want to receive, but isn’t really valuable or something I want at my fingertips” coupled with server-side mail rules and auto-delete cleanup makes my personal email relatively quite and pleasant.

My job is not generally to write code, but sometimes that’s the job. I wish I was better at it. I am so susceptible to the puzzles. I got an itch to deliver something I know is needed at 4pm yesterday. Thought I had it 95% licked by 5pm. But then it turns out the very last step has something devilishly complex.

It’s not an issue of understanding the data or what needs to be done. Instead, I’m in a complex area of code we all desperately know needs to be refactored. I now can choose to spend days in a fraught rewrite, abandon this effort for later, or spend hours making tiny tweaks to find the exact right incantation to pile on more technical debt but get it working.

The problem is I’m a dog with a bone. I thought about it most of the night. I will end up working on it at least a little this weekend. As a quick win, totally worth it. As an all encompassing energy suck— probably not.

Yesterday, after putting away my laundry, Elsa asked me “Did the stain come out?”

I replied, “What stain?” She sighed.

“Last weekend, when I wore your shirt, you were upset because I got some grease on it and you were worried the stain would it come out. Sometimes I wish I had your brain.” She knew I had completely forgotten about the stain and didn’t even check on it.

Then she saw my far off stare. She realized I was already thinking about something else. She said, “Wait, what’s wrong what are you thinking about?”

And I said, “This thing at work I was convinced I understood, but turns out that there’s something hard there I didn’t realize.”

She smiled and said, “I take it back. You can keep your brain.”

I own almost no individual stocks anymore. But I have a little bit of Apple stock, worth about $5,000 that I’ve had in one of my retirement accounts for a long time. I’ve often thought of selling just to get things simpler, but it always performed well and it was such a small portion of my portfolio it didn’t seem worth it.

I came close last week, but reminded myself that my whole investment philosophy (pretty close to straight Boglehead), is that I don’t know anything everyone else who is investing doesn’t know. And without special information, the trick is pretty much hold and stay the course. If you miss just a few of the best days trying to time the market, you can really harm yourself.

So I thought, sure, I have every reason to think these tariffs are real, the recession is coming, the dollar is weak, Apple’s China market is dead, and things will be bad. But everyone knows that, and everyone sold on that. I don’t know anything new, so I’d just be selling low.

Today, it turns out Apple products will pretty much be entirely exempted from the tariffs. I still think they’ve got no more Chinese market and we’re gonna hit a recession. But I bet I’ll be happy on Monday I didn’t sell all that stock at the bottom.

In the end, it won’t matter. But the unpredictable circumstances just reinforce my belief I know nothing and a recession is inevitable.