→ A toolbar that is rudely stamp’d

March 3, 2026
by Riccardo Mori
Handpicked

Nick Heer has published another very good piece, The Window Chrome of Our Discontent, analysing how Apple over the years has progressively done away with the chrome in application windows, in an attempt to prioritise content over the rest of an application’s user interface. A goal that has been repeatedly stated since OS X 10.7 Lion; a redesign whose execution, iterated a few times since then, has been overall increasingly poorer.

Heer shows this in a series of screenshots of Apple Pages’ interface, focusing on how the window chrome and toolbar have changed over time and pointing out what works and what doesn’t. Yes, not everything introduced in a subsequent redesign iteration has been for the worse but, as Heer rightly observes:

Overall, however, what Apple has done to Pages over this period of time is representative of a broader trend of minimizing the delineation of user interface elements from each other and the document itself. This is the only tool in the toolbox, and I am skeptical it achieves what Apple intends.

Compare again the two more recent screenshots against the ones that came before, and focus on the toolbar at the top of each. In the older two, there is a well-defined separation between the toolbar — the window itself — and the document. In the Big Sur visual language, however, the toolbar is the same bright white as the document. By Tahoe and the Liquid Glass language, there is barely a distinction; the buttons simply float over the document. And, bizarrely, that degrades further with the “Reduce Transparency” accessibility preference enabled […]

For me, this means a constant distraction from my document because the whole window has a similar visual language. As the toolbar and its buttons become one with the document, they lose their ability to fade into the background. In the two older examples, the contrast of the well-defined toolbar allows me to treat them as an entirely separate thing I do not need to pay attention to.

Heer’s analysis focuses on the visual aspects of this general regression, and is spot-on. The only thing I feel like adding is that this regression also appears semantical to me (for lack of a better term). Let’s take that last screenshot of Apple Pages’ top window/toolbar and do a quick thought experiment: suppose you’re not familiar with this application. Suppose you’ve never used it before or used it too sporadically to memorise its interface and commands. Look at that toolbar:

 

Can you tell me what all those toolbar buttons do? Can you guess their function by looking at them? Can you guess why some of those controls are grouped and some are not? Can you establish whether there’s a difference in their prominence? I showed this toolbar to a couple of people who primarily use Windows, and they were both fairly puzzled.

What’s the difference between (1) and (2)? They both seem to add or insert something, but what exactly? What does (3) do? Or (5) for that matter. Or (6). Why are (4) and (6) in a different colour than the rest of the buttons? The icons for (7), (8), and (9) are still familiar enough, so one can easily guess they are used to insert a table, a graph, and some shape, respectively. (10) is already more ambiguous. If this were an email client, this would be the button to add an attachment to an email message, but this is not an email client. Could it still be referring to some kind of attachment? (11) too is a bit vague, but you can ultimately guess it’s about inserting a comment… Not to open a chat window between collaborators, right? (13) is misleading if you don’t already know Pages. A paintbrush? Could it refer to pasting something? But if it were a Paste command, where are Cut and Copy? And is (14) used to add a page or change the page layout? Maybe?

Toolbars like this work much better with an icons + text view and a much clearer distinction between their function, if they’re direct commands or if they just open inspectors or additional tools. I don’t have Mac OS 26 installed, so I don’t know whether Pages’ toolbar now comes in icon view only or not. If it does, then this is a rather cryptic toolbar for novice users. If you design for purely icon-based toolbars, then you must do additional work to make those icons as clear and unambiguous as possible. If your toolbar has multiple views and can be displayed as icons + text or even text-only buttons, then you can afford to have a couple of icons that aren’t very immediate at first glance, because when you have such multiple view options in toolbars (as Mac OS historically has had), they usually default to an icons + text view. Once you got familiar with what all the buttons do, you can save some application window real estate by switching to an icon-only view.

And what’s amazing is that, if you go back to Heer’s article and look at Pages’ toolbar under OS X Lion, all those icons are rather self-explanatory when you look at them without their labels. That happens thanks to the icons being more colourful, detailed and descriptive. Of course the Comment button is a Post-It note. Of course the Media button is instantly recognisable. Of course all the buttons that open a menu instead of triggering a direct action do feature a small expansion triangle next to them. They all have these little visual cues that you consistently see in other places of the operating system’s UI. But this, of course, only works if you have a consistent operating system’s UI, and not a patchwork of different ideas assembled by different committees.

→ On software quality (and reliability, and frugality)

February 25, 2026
by Riccardo Mori
Handpicked

On Software Quality is a very fine article by Nick Heer I suggest you read. There’s a passage I don’t fully agree with:

There was a time when remaining on an older major version of an operating system or some piece of software meant you traded the excitement of new features for the predictability of stability. That trade-off no longer exists; software-as-a-service means an older version is just old, not necessarily more reliable. 

I have several older Macs I still use, some on a daily basis (like my 2017 21.5‑inch 4K iMac and my 2010 17-inch MacBook Pro), some on a frequent-enough basis (like my 2013 11-inch MacBook Air and my 2015 13-inch MacBook Pro), some others just occasionally (like my 2008 13-inch black MacBook or my 2009 15-inch MacBook Pro). The two main factors that contribute to their obsolescence are:

  1. Progressive unavailability of up-to-date Web browsers (and up-to-date digital certificates to guarantee online security).
  2. Cloud services like Dropbox and OneDrive progressively dropping support for older Mac OS versions.

Typically, for how I use all these computers, №1 is more disruptive than №2. I very much enjoy using older Mac OS versions, but not being able to browse the Web properly and securely, not being able to correctly sign in to check a Gmail account, not being able to fetch some RSS feeds because you can’t authenticate securely or establish a secure connection is very frustrating. Not having Dropbox work on my 2009 MacBook Pro running OS X 10.11 El Capitan is a minor annoyance and means I just won’t have access to certain personal files and that I’ll have to sync manually whatever I do on this other machine.

But if I put these two factors aside, there’s nothing about those older Macs, nothing about the older Mac OS versions they run that makes them less reliable. The crystallisation of the operating system they use and the software environment I find on them is exactly what makes them more reliable than the newer stuff. Just because an application has been discontinued by Apple — like Aperture — doesn’t mean it has stopped working or has stopped being reliable. Just because a third-party app has moved on from supporting a Mac OS version (or even a whole Mac architecture) doesn’t mean I can’t keep using the previous version of such application with that older Mac OS version. I’m aware this isn’t exactly what Heer was arguing — I’m just saying that when I use these older Macs with older Mac OS versions, it’s like entering a snowglobe-like environment where everything that still works by current standards or demands, still works reliably and predictably. And what doesn’t work, well… just doesn’t work. It really doesn’t make the Mac or its older Mac OS version less reliable.

Now I’ll admit, one important thing that works to my advantage is that I have been using for years more or less the same bunch of core third-party tools made by Mac developers who:

  • have released either free software or software that can be purchased upfront without a subscription; or even software that has a usable free tier and requires a subscription only for its more ‘pro’ features (BBEdit is a perfect example of this);
  • have maintained access to older versions of their software applications, making said versions free to use or unlockable by providing a valid licence for a more up-to-date version of their application.

So yes, I’ve tailored my software experience around predictability and stability for quite a while now, favouring these aspects over ‘needing’ to update just because a piece of software promised some fancy new features. The proliferation of subscription-based software — or software as a service, as Heer says — has clearly made things a bit trickier in recent times. If you, unlike me, heavily rely on subscription-based apps for work and leisure, then yes, remaining on an older major version of an operating system or some piece of software can present an issue and may end up being unwise in the long run. But I wouldn’t frame the issue in terms of ‘reliability’ — more in terms of ‘availability’ or ‘compatibility’.

I think the trade-off of staying on an older major version of an operating system or third-party application to pursue stability instead of buying into an increasingly mindless update cycle still exists. And that while an older software (or system software) version may not necessarily be more reliable than a newer one, it doesn’t mean it is necessarily less reliable or stops being reliable altogether.

Nick Heer concludes:

What I expect out of the software I use is a level of quality I simply do not see. I do not think I have a very high bar. The bugs in the big paragraph above are not preferences or odd use cases. They are problems with the fundamentals of the operating system and first-party apps. I do not have unreasonable expectations for how things should work, only that they ought to work as described and marketed. But complaints of this sort have echoed for over a decade and it seems to me that many core issues remain unaddressed.

People buy hardware, and it shows. People subscribe to services. But people use software. This is not solely an Apple problem. Many of us spend our time fighting with tools that feel unfinished and flawed; it seems to have become the norm. But it is particularly glaring when the same attitude is taken by Apple, a company that ships some of the nicest hardware in the business. I would love to see the same tolerances for what is shown onscreen as Apple has for how the screen is made. 

What’s really sad in all this is that many of those “problems with the fundamentals of the operating system and first-party apps” aren’t structural; that is, they’re not derived from historical faults or shortcomings in the fundamentals of the operating system. They often are the result of more recent bugs breaking something that used to work or a solution that had already been found, and said bugs have been allowed to fester thanks to an unsustainable yearly release cycle that forces engineers to work on new features instead of fixing what broke down in previous iterations. This core issue in software’s ‘cardiovascular system’ is equally felt at the major arteries’ level and at the capillary level, which is that ‘fighting with tools that feel unfinished and flawed’ that Heer talks about.

By the way, the ‘software as a service’ model, in this scenario, does nothing but exacerbate these functional issues. The perennial, flawed, beta state of software is maintained at the OS level because this system software — instead of being designed to achieve a final, self-supporting state — is designed to be continually patched, retouched, refined on an asymptotic line (a process that creates a lot of baggage and technical debt). And when it comes to third-party apps, having software that is considered a service and not a product means, once again, that it is not designed to achieve finality, but to remain available in a usable state for an indefinite amount of time. We can debate exceptions, good intentions, and finer points ad nauseam, but I maintain that this progressive departure from viewing and structuring software as a product has been severely detrimental to the nature, quality, and usefulness of software applications for the end user. 

It has disincentivised a lot of big and small companies from striving for excellence and releasing truly great products, instead opting for software that is decent enough but always has some room for improvement, a gap that is never really filled because otherwise the updates would stop coming and it wouldn’t make sense to pay a subscription fee anymore. In the best-case scenario you may have developers who genuinely strive for releasing great apps but their work is disrupted by Apple’s update cycles, forcing them to rewrite code and update their apps so that they keep working after a major Mac OS release. This is why I was using the ‘cardiovascular system’ metaphor before: the flow at the operating system level (major artery) affects the flow at the third-party app level (the capillaries). I really hate this kind of entropy because it feels largely avoidable. But reversing the course needs significant efforts, and such efforts have to come from Apple; they theoretically have the resources to shoulder these efforts and lead by example. I’m not holding my breath.

On software frugality

When technology was kinder to its users, and when someone like Steve Jobs was still alive and had Mac users’ interests at heart, keeping my Macs updated was a pleasure. When a new Mac OS X version was presented, I knew Apple had been at work to genuinely improve on the previous version, so for me updating was a matter of when more than if. And up until probably Mac OS X 10.8 Mountain Lion, ‘when’ usually meant ‘as soon as possible’. 

But then, the ‘new cool features’ promised by the newer Mac OS version started feeling less important or groundbreaking. More cosmetic. Or more tied to a vision of the ‘Mac experience’ that has progressively felt more detached from the regular user’s reality, concocted by executives from their ivory towers at Apple Park who don’t seem to actually know how people work with their Macs in their day-to-day.

So there was a moment where upgrading to the next version of Mac OS turned into something I did reluctantly, often waiting two or three minor version updates because the .0 releases were getting buggier and broke more things. I even skipped OS X 10.10 Yosemite entirely because the visual changes were just too jarring for me and Neue Helvetica as new system font was really working against my eyesight — and against common sense, from a design/UX standpoint. OS X 10.11 El Capitan felt better in many departments, so I jumped back on the update bandwagon.

Anyway, to make a long story short, over the past 10–15 years my attitude towards Mac OS and software in general has shifted and has become more self-centred. I ask myself more questions. What can this new update do for me? Is there anything that the previous iteration can’t keep doing? No? Then what’s the point of upgrading? Does this app have some groundbreaking feature I was really looking for or missing from my tools that makes it worthwhile to start a subscription and rent software I’d really prefer purchasing? No? Then I don’t need this app.

Today more than ever, technology wants to put you in a river where you flow from update to update unquestioningly; a river where you keep flowing forward because the ‘new’ is always better than the ‘old’. I went along with this until it stopped ringing true. Today more than ever, technology and tech companies feel like entities that don’t work for us and don’t have our interests at heart; they just want us to depend on them utterly and continually. So I have to look out for my needs.

That means making difficult choices sometimes. That means questioning convenience (convenience for whom?) and reintroducing friction. I can’t use the term minimalism with a straight face anymore, that’s why I prefer frugality. I can’t provide a set of ‘rules’ or a list of tips & tricks here. This is the classic situation where your mileage really may vary a lot. What I can do is share an attitude. You should have already gleaned it from the previous paragraphs, but I’ll reiterate.

Put yourself and your needs first. Anything that works against that is not worth your time, energy, money, or obsession over it. A piece of software that is ‘nice to have’ but in exchange keeps you hostage through some form of lock-in or yet another subscription that further erodes your budget (especially if you don’t have money to burn)… is it really nice to have? An operating system that forces you to adjust your habits and workflows on a yearly basis… what does it really have to offer that makes all that hassle worthwhile? That new app that does the same thing as that old app you trusted… do you switch to it because it’s actually better or just because it’s new and ‘looks fresher’? If upgrading to Mac OS 26 means you have to also update a bunch of apps and all these now have worse usability — because Mac OS 26 is the worst usability regression in all Mac OS history — is that upgrade working for you or against you? If you reach a point where the entire software ecosystem you’re using is constantly thwarting you in a big way or even in a myriad of smaller ways, where do you draw the line? Do you even draw a line? Is the constant acceptance of the compromise of convenience, still convenient?

When you think in terms of ‘your needs first’, in terms of ’none of these tech companies is your friend’, and you build your software toolbox around the concept of ‘essential tools’, migrating to another platform isn’t as daunting or traumatic as it feels when you start imagining it. You do a bit of homework to prepare yourself. You ask yourself what kind of applications you need to look for if you change platforms, and evaluate the best candidates. Sometimes you get lucky and you find that the same company that made one of your favourite tools has also made a version of such tool for Windows or Linux. In this case you don’t even have to adjust to the new tool. In other cases, you might have to do additional work to become as productive in the new platform as you were in the old. But if you’ve answered ‘Yes’ to the question ‘Is it time to leave this platform behind because it takes from me more than it gives me?’ then you’re willing to do the additional work of forming new habits.

Where am I now in all this? I’m at a point where I’m not upgrading to Mac OS 26 or any other 26 software release from Apple. If the poor state of Mac OS user interface, user interaction, and usability is not radically addressed in future versions, I will keep using older Mac OS versions until the resulting friction is no longer sustainable. My main Mac is still on Mac OS 13 Ventura, and I’m planning to upgrade to Sonoma very soon. Considering how I use my Mac and my tools, the impact of staying on this older version of Mac OS has been non-existent so far. Everything I need to work still works without issues, and I’m still enjoying the benefit of working with a Mac OS version that — while not as perfect as Snow Leopard, or as stable as El Capitan, High Sierra, and Mojave still feel today — is way, way more legible, usable, and ‘out of the way’ than Mac OS 26 with its Liquid Glass UI. 

I’m already using Windows and Linux as secondary platforms on other hardware. I don’t think I’ll ever remove any trace of Mac software and hardware from my daily life. The beauty of not needing ultra-specialised software that has to be constantly kept up-to-date means that I’ll be able to hold on to older, trusted Mac apps for a long time. But if I ever need to move away from the Mac as my main computing platform, switching to Windows or Linux won’t be much of a culture shock as it often felt in the past. And if my day job, as I fear, dictates that I’ll have to keep upgrading to newer, worse Mac OS versions, then I’ll make sure to do that on a dedicated machine where I don’t do anything else but work.

People and resources added to my reading list in 2025

February 13, 2026
by Riccardo Mori
Tech Life

Welcome to the thirteenth instalment of my annual overview of my most interesting discoveries made during the previous year. A few months ago, a friend of mine remarked that the title of this series of posts should be updated, because instalment after instalment, my list of things to actually read has become shorter, and the list of resources to watch has become longer. Maybe you should just say, “People and resources added to my watch list”, they suggested. But ‘watch list’ gives me bad surveillance vibes, and discovering and suggesting new blogs always has priority for me, so ‘reading list’ it is. Or perhaps I should go back to the wording of the first post of the series, published in early 2013 — Some interesting resources I discovered in [year]. We’ll see.

Apologies for the slightly navel-gazing introduction. 2025 was another ‘difficult’ year if you hadn’t guessed from articles like My 2025 in review, and Not fatigue, but disconnection; a year where I began to revisit older habits that used to stimulate me more and kept me from looking at screens all the time. Among these, reading physical books and engaging in more active music listening by actually sitting in my other studio (where I keep the bulk of my library and my hi-fi stereo), and listening to whole albums while keeping a notebook handy in case this activity triggered some new ideas or inspiration for my creative writing.

This and the promise I made to myself to be more selective in what I actually decide to add to my repository of resources-worth-keeping, resulted in yet another short overview.

Blogs

  • The website of V.H. Belvadi. I discovered Mr Belvadi after an acquaintance passed me a link to this piece, The death of an argument — Why analogies are best used sparingly. I enjoyed it and liked his analysis, so after reading I started exploring his site. As you may recall (or not), in last year’s instalment I wrote, You know what happens when you get even more selective? That maybe you follow a link to a blog article, and you like the article, but then you explore that blog further and you realise that such article — and perhaps a couple more — is the only highlight of that blog, and you start wondering, “Is this website worth adding to my RSS feeds, or should I just share the link to that specific article and let others decide?” In most cases, I’ve ended up bookmarking & sharing articles instead of adding blogs to my reading list. This was definitely not the case with V.H. Belvadi. He writes consistently well, and consistently interestingly about a varied range of topics. I encourage you to check out his website and add him to your RSS feeds.
  • The website of Jason Velazquez. It all began with someone I follow on Mastodon, who boosted a post by Jason I enthusiastically agreed with. I’m a curious person, so first I checked Jason’s profile to have a general feel of the kind of things he posts about. I kept liking what I saw, so the next step was to visit his website, read the article referenced in one of his most recent Mastodon posts (Hank Green And The Fantastical Tales of God AIs), and continue to read other pieces by Jason. I invite you to read that article and then explore Jason’s website, which is a delight to navigate. Jason’s writing is meaty, evocative, pragmatic, and I know these adjectives may feel a bit contradictory — I’ll let you visit his site and see for yourselves.

What these two websites have in common is that they’re designed to look, feel, and be navigated like books. I’m a bit jealous of such designs, because that’s how I always viewed my own website, but my limited coding knowledge has always prevented me from reaching these lovely results. (Suggestions of using ‘AI’ tools for this purpose will lead to excommunication; you’ve been warned).

YouTube channels

After last year’s intervention, the situation with my excessive number of YouTube channel subscriptions has normalised and returned to healthier numbers. As you’ll see below, I did indeed add a dozen new subscriptions, but a lot of the following channels have a somewhat relaxed publishing schedule, so things rarely get overwhelming.

Gaming-related

  • Riloe and Ratat are two creators whose channels focus mostly on gaming essays. Riloe’s essays are more about tactical and extraction shooters, while Ratat talks more about horror games, and I find his analyses of Supermassive Games’ Dark Picture Anthology games to be well worth a watch.
  • Euro Brady — Brady is a therapist, and provides a very interesting perspective and a fresh angle to the usual ‘let’s play’ style of gaming videos. Unlike so many other letsplayers, Brady doesn’t rush through game levels, lore, etc. but instead frequently stops and analyses the personality and psychology of the various game characters. What’s really great, in my opinion, is that talking about the character’s psychology or struggle before a certain situation in the game isn’t an end in itself, but a starting point to talk about mental health in general. Brady’s digressions have often helped me understand certain behaviours in real life, certain interpersonal dynamics and interactions we may find ourselves in. Follow him if you appreciate this kind of insights, more than watching someone reach the end of a game. The game is not really the point here.
  • nocaps — Indie games reviews and overviews. I really like her calm and pleasant personality.
  • itsTedBrooks — Ted has a small channel that started with filmmaking-related content and turned its focus to gaming essays over the past year. His videos are on the short side, well scripted, edited, and shot. I discovered him thanks to YouTube’s algorithm, which one day suggested I watch We don’t play games anymore… We just argue about them. I turn that suggestion to you. If you like it, then watch Ted’s video about the game CONTROL, and then subscribe!

Tech-related

Two very different approaches to technology here:

  • Janus Cycle — The best thing to introduce you to this channel is its very own description: Exploring retro devices from the pockets of history. Janus Cycle is an eclectic journey into technological wonders of the ages. Follow along with detailed looks into a wide range of intriguing devices, while appreciating the marvels of technological miniaturization. Going deep into the technological aspects that makes each device unique. Following their stories into inventive and sometimes esoteric ways they function or affected our lives. Old mobile phones, PDAs, computers, assorted devices… This is a channel for the tinkerer and the person interested in discovering more about the technology behind many devices from the past 2–3 decades. A real gem, in my opinion.
  • Our Own Devices — From the channel description: Our Own Devices is a channel dedicated to the fascinating world of vintage technology, and the many elegant and ingenious ways our ancestors solved even the most complex technical problems. [Update: YouTube has just removed this channel because “it violated our Community Guidelines”. I wonder how, given that the channel was a fascinating exploration into the workings of devices such as mimeographs, portable record-players, radios, and many many other items from the 20th century. Update 2: The channel has been reinstated. It turns out it had been hacked.]

Other

  • Voynich Talk — I’m obsessed with the Voynich Manuscript. This is probably one of the best YouTube channels about it.
  • The Late Late Horror Show — Mainly movie reviews and nightly streams of old time radio shows.
  • Quarantine Collective — From the channel description: The Quarantine Collective is the home for a new kind of philosophical pedagogy. Instead of experts telling you how things work, we encourage discussion, transversality, and rely on the participation of every skill level. Join a live stream as we make our way through texts collectively, with the support of experts, academics, and hobbyists alike. The main force behind the channel is Brooks Brown, and the channel isn’t exclusively focused on philosophy. Many of the Quarantine Collective’s ‘After Hours’ streams are more free-form, with the host reacting to other YouTube videos and bringing his wealth of knowledge and common sense to the table. He has also made one of my favourite video essays on LLMs and ‘artificial intelligence’: No, AI is not Sentient (It’s just more Capitalism)
  • Anna Bocca — Anna’s video essays are mainly focused on economy and society. She is a great communicator and I really like how she edits her videos, patiently crafting the visuals and infographics. Me and economics are like water and oil, and she managed to make me understand a few things that normally would have flown over my head…
  • baby.murcielaga — Fantastic music compilations, mainly ambient and vaporwave, but not limited to that. I have no particular favourite to recommend, just dive in and explore.
  • Chris and Jack — They are an amazing duo making comedy sketches that feel more like short films. Great humour, great scripts, and very well produced material. My first exposure to them was this: Sci-Fi Movies never pick the right year. You’re welcome.

Podcasts

Another year, another round of copying-and-pasting the same quote from a few years ago:

In 2019 I unsubscribed from all the podcasts I was following, and I haven’t looked back. I know and respect many people who use podcasts as their main medium for expression. My moving away from podcasts is simply a pragmatic decision — I just don’t have the time for everything. I still listen to the odd episode, especially if it comes recommended by people I trust. You can find a more articulate observation on podcasts in my People and resources added to my reading list in 2019.

If you’re wondering why I keep the Podcast section in these overviews when I clearly have nothing to talk about, it’s because to this day I receive emails from people un-ironically asking me for podcast recommendations.

Useful Web tools

  • PiliApp — A collection of fun and cool Web tools.
  • ColorPalette Pro — Load it in your browser. Play around. Yes, it’s about colour palettes.

My RSS management

Yet again, nothing new to report on this front. I’m still using the same apps I’ve been using on all my devices for the past several years, and I haven’t found better RSS management tools / apps / services worth switching to. In my previous overviews, I used to list here all the apps I typically use to read feeds on my numerous devices, but ever since I broke my habit of obsessively reading feeds everywhere on whatever device, I’ll only list the apps on the devices I’ve used over the past year or so. If you’re curious to read the complete rundown, check past entries (see links at the bottom of this article):

  • On my M2 Pro Mac mini running Mac OS 13 Ventura: NetNewsWire.
  • On my 17-inch MacBook Pro running Mac OS 10.14 Mojave, and on my 13-inch retina MacBook Pro running Mac OS 11 Big Sur: NetNewsWire 5.0.4 — A slightly older version of this great RSS reader.
  • On my other Intel Macs running Mac OS 10.13 High Sierra: Reeder and ReadKit.
  • On my iPad 8: Unread, Reeder, NetNewsWire for iOS, and ReadKit.
  • On my Android phones — Nothing Phone 2a and Microsoft Surface Duo: the Feedly app.
  • On my iPhone SE 3, iPhone 8, iPhone 7 Plus, iPhone 5s, iPhone 5, iPad 3: Unread. (Though on the iPad 3 Reeder seems to be more stable and less resource-hungry).

Past articles

In reverse chronological order:

I hope this series and my observations can be useful to you. Also, keep in mind that some links in these past articles may now be broken. And as always, if you think I’m missing out on some good writing or other kind of resource you believe might be of interest to me, let me know via email, Mastodon, or Bluesky. Thanks for reading!

→ The real AI problem no one is talking about

January 30, 2026
by Riccardo Mori
Handpicked

The other day, YouTube’s algorithm struck again. It suggested a video podcast episode from a creator with a still small channel, Josh Allan Dykstra. The title, while sounding moderately clickbait‑y, still made me curious to check out the video:

The Real A.I. Problem No One Is Talking About (Who Buys Your Stuff, Robots?)

I’m always interested in intelligent analysis debate around ‘AI’ topics, and Dykstra isn’t someone who’s profiting from ‘AI’, so his analysis doesn’t look biased to me.

Conveniently, Dykstra also provides transcripts of his podcast episodes on his site. Here’s the one for this video.

He poses the question right at the start:

Today, let’s start with the wrong question.

Everyone keeps asking: “Will A.I. take our jobs?”

I get it. That question is terrifying because it’s personal. But it’s also incomplete.

There’s a better question than if A.I. will take jobs — because, spoiler: it will, and it’s already happening (you’re noticing the hiring slowdown, right? [in the US] 2025 was the worst year for hiring since 2009, with the exception of COVID year 2020).

I don’t want to minimize the impact of A.I. taking jobs (it’s just beginning and it’s going to be enormously disruptive) AND there’s actually a bigger question lurking in the background, namely: What kind of economic system tries to eliminate human labor without replacing income… and still expects the world to keep functioning as-is?

I’m not hearing this talked about enough, so we’re going to talk about it today.

Here’s the crux of it: if A.I. works the way capital hopes it will, capitalism won’t break because A.I. fails. It’ll break because A.I. succeeds.

Why?

Because if you’re running a business, you can’t fire your customers and then expect them to buy your stuff.

There’s a simple loop that runs the modern economy: Labor → wages → income → consumption → revenue… and then back around again.

A company pays labor (that’s you) in wages which become your income. You have some left over so you buy things (consumption) which is another company’s revenue, which they use to pay their labor. Feels familiar, right?

In the way that modern life currently works, this loop is NOT optional. It’s what we might call “load-bearing” — it’s holding up the house. We break or pull out one part of the loop and things will NOT function the way they do now. Period.

[…]

Losing our jobs wouldn’t be so scary, of course, if we had another way to get income. The problem is most of us don’t. Our labor is what we trade to get to income. […]

As in all technological disruptions, capital is going to use A.I. to eliminate labor because labor is… inconvenient. 

I get it — we humans are dreadfully biological — but without labor people don’t have income which means they can’t participate in consumption… which means Capital is also automating itself out of relevance. Because this time, the goal of the technology is quite literally to do everything a human can do (this is probably the most generally accepted definition of AGI). 

We all see the problem, right? If no one has any money, who actually buys the stuff your robots make, Capital?

The strange irony here is that this happening wouldn’t actually be capitalism failing, it’d more be like capitalism completing its own logic — like the snake that kills itself by eating its own tail. 

I’ve tried to summarise Dykstra’s argument the best I could, but I suggest you watch the whole podcast episode (it’s just 20 minutes long) or read the whole transcript. The question he poses is something I’ve been pondering myself ever since this ‘AI’ craze began to propagate. 

I’m still not sure whether the doom-and-gloom scenario of ‘AI’ is coming for our jobs is really going to materialise in full-dystopian mode, bringing that high level of disruption Dykstra talks about. But even if we talk in hypotheticals, this is a problem worth considering. If entire categories of workers lose their job, they stop gaining money. If they stop gaining money, they stop spending money, and that means that other people or companies will stop earning revenue. That whole loop Labour → Wages → Income → Consumption → Revenue → Labour, etc. is going to fall apart. 

I can’t wait to hear the ‘solution’ from some sociopath techbro.

The digicams return

January 22, 2026
by Riccardo Mori
Briefly

Elephant Memory Systems entry №002

In recent years, the term digicam has come to indicate more than just ‘a digital camera’, as it was first used to refer to digital cameras as opposed to film cameras. Now the term also references a specific type of device: compact digital cameras produced from the late 1990s to the late 2000s, typically using older CCD sensors. This has been fuelled by a general rediscovery of the vintage, imperfect, more organic look of the photos these cameras produce.

As someone who has been into photography since the 1980s, and who has used these compact cameras when they were the latest and greatest in digital photography, this ongoing trend is more than just a fad or just another excuse to achieve originality or to look, well, trendy. And it isn’t fake nostalgia either. This type of photography brings back real memories.

The serendipity of rediscovering digicams and the digicam æsthetic as a personal endeavour slightly before this became a trend on Instagram, YouTube, and elsewhere is what essentially spared me from wasting a lot of money on 20-year-old cameras that should cost just a few bucks second-hand, but whose prices have been horribly inflated thanks to a bunch of photography ‘influencers’ babbling about ‘the film look’ these old, little CCD cameras supposedly give you.

I currently have more than 30 of these digicams, and I had a lot of fun taking street snaps around the city between 2021 and 2024. In mid-2024, however, my wife and I finally bought an apartment and for a few months were busy with everything related to buying and moving to a new home. In short, my digicam collection has mostly lain neglected for most of 2025. Further, last year my creativity in general took a hit also because of the stuff I talked about in My 2025 in review.

But during the Christmas holidays, as I finally finished reorganising hundreds of family slides to show more about my family and childhood to my current family (my wife, and her brother and sister), a renewed yearning for getting back to photography as a whole, and for dusting off my digicams again, returned in full. These days I’ve been busy recharging many camera batteries and taking out a couple of digicams each day with me as I went out and about.

The first two cameras I revived have been the Sony Cyber-shot DSC-L1 and the Olympus FE-250. The DSC-L1 is a very small, mostly metal, camera, that is rather comfortable to use despite its size. It has 4.1 megapixels and was made in 2004. The Olympus, another very pocketable camera, is from 2007 and has 8 megapixels instead.

I tried to capture the same scene with both cameras, but honestly, this is not a great sample of what these cameras can do.

Sony Cyber-shot DSC-L1

Olympus FE-250

The Sony DSC-L1 is rather capable at handling difficult light (a mix of natural and artificial, like in the first photo) and has a good auto white balance (the lighting in the second photo looks very similar to what I was seeing).

 

 

The Olympus FE-250, like all the (many) Olympus cameras I own, delivers very nice colours.

 

 

The second pair of cameras I took out shooting have been the Sony Cyber-shot DSC-W370 (14.1 megapixels, 2010 — above) and the Sony Cyber-shot DSC-T77 (10.1 megapixels, 2008 — below).

 

That plastic tab attached to the T77’s strap is in fact a mini stylus, as the T77’s back is entirely taken up by a 3‑inch 16:9 capacitive touchscreen. The only physical controls are the shutter button, power button (you also turn the camera off/on by sliding the front panel), review button and zoom rocker.

Sony Cyber-shot DSC-W370

Sony Cyber-shot DSC-T77

The DSC-W370 cost me almost nothing, as it was part of a box of 12 assorted cameras I got on eBay for €30 or so, sold as ‘untested, spares or repairs’ — and 11 of them turned out to be working fine, either needing new batteries or chargers. This Sony just needed its charger.

The DSC-T77 was purchased at a second-hand shop for the princely sum of €12. Both purchases were made in 2021. This camera was originally available in a few colours. The promo image gives iPod vibes (credit: DPReview).

It comes in colours…

At the moment I have the third pair of digicams in my bag: the Canon PowerShot G2 (4 megapixels, 2001) and the Fujifilm FinePix E550 (Super CCD HR sensor with 6.3 megapixels, 2004). Both these cameras have taken many very good pictures in the past, and it’s been great getting back to them these days. Maybe I’ll show more sample photos in a future ‘Elephant Memory Systems’ entry.

 

See the EMS tag for more short-form posts of this kind. Read this entry for the origin story of this series of posts.