Where Did Personal Computing Go Wrong (2020)
mmcthrow-musings.blogspot.com
mmcthrow-musings.blogspot.com
This is the most telling quote in the article. Companies took over computing by locking down Cell Phones. Now this model seems to be in the process of moving into to M/S Windows 10/11. I do not use MAC/OS, but to get your application in the MAC store, do you need to pay a fee or give a cut if you sell it ? I suspect you do, or where else would Microsoft get their Ideas from.
In anycase, to me if you are not using Linux or a BSD, I think you are slowly being moved into a Walled Garden. I also noticed in Linux there is now a "GNOME Store" and a "KDE Store", so far, I think those still remain free for developers to get their work into.
But seems to me, Linux is very slowly becoming commercialized, who knows where Linux will be in 10 to 20 years. I am starting to think eventually the only place people in the future will have access tp Linux is via WSL. With Secure Boot becoming more accepted, I think that is a very big risk.
It's interesting to remember why this was possible/accepted. As I recall, at the time of its release, the iPhone's App Store was a breath of fresh air compared to software stores on traditional cell phones that were often tightly controlled by carriers and featured prices that were outrageous even by 2007 standards. For example, I remember a very crappy breakout clone being sold for ~$20 on my Nokia flip phone.
So in a weird way, cell phones weren't "locked down", they were "freed up" relative to what had come before. However, as cell phones became the primary computing platform for a generation of people the restrictions didn't lift to the point where cell phones were on par with traditional PCs and instead we've been stuck with this "good enough" situation of a mostly locked down platform with a few workarounds (side loading APKs, TestFlight) for people that really need them.
I blame a lack of competition. There has been no viable competitor to the Android/iOS duopoly since Microsoft gave up on the Windows phone. And certainly no one with the marketing chops to turn a technological advantage like "open platform" into a killer feature that would get the masses to buy in.
Then Google finished the job; side-loading was no longer "jailbreaking", it was directly supported, and programming on the device wasn't prohibited. I'm not the customer for an iPhone, I really doubt they'd ever open up so that iPhone owners could install an open source app store (f-droid), and from that a full development environment (termux), within which you can install a bigger distro (Debian) to install e.g. Calibre for ebook-reformatting. Or git, or LaTeX, or rust, or emacs.
My first streaming setup basically used SSH to control my MPD server but it worked.
Buying a product from the manufacturer and selling it at a markup to end-users is called "retail sales" and is a concept that predates software by a few years.
You do not need to put your app on the store. Macs come free with Xcode and always have since OS X was released and you can compile and run whatever you want. You can sign code without submitting to the App Store if you really care.
Macs are friendly to normies, but that doesn't mean you can't tinker with them. You can. They are like an iOS device and a UNIX system and everything in between, not only one or the other.
The day they start making you jailbreak a Mac is the day I move to linux, but until then, most of what you hear about them are just bs from people who've never used one.
If number of potential users of alternative operating system is less than 1 percent, then that hardware will never be produced.
I am in the process of moving all of my machines away from Linux and to BSD as a result of this trend.
I think if Linux refused to support secure boot and paying M/S for keys, I think we would be in a better place.
https://www.omglinux.com/boot-linux-modern-lenovo-thinkpads-...
Walled Garden is marketing speak. "Closed Ecosystem" is more objective(neutral) and doesn't have positive words associated with it.
We should avoid using marketing speak.
Because it does.
Politicians and marketers all dress up their words.
If it didn't we wouldnt have marketing departments or speech writers. Are you denying word choice doesnt matter at all?
Garden is clearly a happier, more inviting, prettier, and nicer place to be than an ecosystem. A garden is in your backyard at home, full of healthy fruits and pretty flowers. Ecosystem? Could be a petri disk of bacteria or the entire earth. Heck, you can google/chatgpt the definition of these two and find out which has a more positive perception.
We are animals at our core, garden drives warm emotion, ecosystem doesnt.
To me, the meaningful part of "walled garden" isn't the "garden" part, it's the "walled" part. It's like "gilded cage" -- the important part of that is "cage", not "covered in gold". "Gilded", like "garden", in these contexts, implies deception to me.
But even so, "ecosystem" has very positive connotations to me, no less positive than "garden".
The story of Microsoft’s late 80s and 90s dominance is fading fast.
My kids, now in the early to mid-teens, have grown up with Apple as the dominant phone, tablet and watch platform. All their friends have them. Here in the US (NY metro), Android seems a second rung system. Common but not poplar.
They use Chromebooks and GSuite at school.
Windows is an after thought. Even on the cheap Windows laptops we have for them at home they’re not really “Microsoft” machines. No Office, no heavy O365 anything. They’re terminals for Chrome and web-apps.
The rest of the article is interesting, but user-developed improvements can be found in fun and interesting places.
The same kids play Minecraft and Roblox heavily. My 12yo has made and sold Roblox clothing. It’s not “programming” or self-solving problems, but it’s interesting creation in a similar vein. Minecraft is full of people creating fantastic and interesting mods and improvements to it.
And for premium phones, iOS has had >50% market share for a long time: https://www.counterpointresearch.com/apple-takes-62-premium-...
* University is switching to Microsoft from Google for mail/calendar
* Some of us use Powerpoint for presentations, and a few docs are sometimes written in word
* We certainly use Github a lot
* We advertise/promote on LinkedIn
* We sometimes take courses through LinkedIn Learning (paid by the university)
* Now we could collaborate with them (?)
* I use a Surface Pro
Despite us being very heavy Linux users (like a lot of science), Microsoft is still everywhere.
MS is still around, and huge, but in a less obviously pervasive and all dominant way.
LinkedIn isn't branded as MS, and I don't remember even seeing it mentioned anywhere. Same with Github.
Contrasting with me original post about the MS of the 90s where there were few to no alternatives, and everything they produced was part of the one monolithic Wintel PC, now they're distributed and far more open. I know there's a worry about linux getting popular on Azure and then somehow screwed up by MS, I think that their quiet distribution is a sign they're less all powerful.
Github users could move to Gitlab, Bitbucket (sort of), or self-host much easier than moving off MS used to be.
LinkedIn is far less of "the way you use a PC" than any old MS product, and has a pretty specific niche market. It's huge, but it's not the invincibly moated. Socials come and go all the time. There are lots of LinkedInU alternatives too.
I don't mean to say MS aren't huge or pervasive - they are. We're just so far on from their absolute dominance of all tech that people coming up on adulthood have no idea the world used to be MS or nothing. (For a given value of nothing, as a 90's Mac user).
Apple exploits in-group bias among teens. This is less of a 'we moved to a closed ecosystem' and more of a 'my kids do whats kool among teenagers'. At the end of the day, the hardware and software are near clones of Androids. This is a move away from computers sure, but it doesnt have to do with closed stores. It has to do with insecure teenagers trying to be popular.
> Even on the cheap Windows laptops we have for them at home they’re not really “Microsoft” machines. No Office, no heavy O365 anything. They’re terminals for Chrome and web-apps.
But that is often how it starts. One day you need to run something locally, and you learn how the OS works. You learn about folder/file structures. You edit settings, you might even need to upgrade hardware.
At worst I can see the future be phones/tablets/laptops logging into an external server to do 'computing'. You still will be accessing programs and data on the computers with lots of control. Isnt that the point?
At first, the "Windows is an afterthought" comment made me squirm inside, as in order to game or do anything remotely hardware intensive one would have to move over to a self-built machine, which means leveraging Windows or Linux.
However, it does look like we are (perhaps not so slowly) moving towards an always-online world, in which everyone owns a dumb terminal to which they access powerful hardware on the cloud. Want to game? You can do so on NVidia's or Microsoft's hardware. Want to run a complex machine learning algorithm? Huggingface has got your back. I fear that's where we're moving toward, and in my opinion, takes away freedom from the end user.
And so many stakeholders are aligned towards it. Dumb terminals that sign in to a persistent account are easy and cheap for users. Companies are in control, enjoying DRM and analytics like never before. Governments can surveil their citizens more deeply than ever by subpoena of the cloud company.
But there's some hopium that, with things like homomorphic encryption, you could have the best of both worlds: cloud connectivity and privacy.
1) WAY better than a closed walled garden like e.g. Nintendo and also
2) Completely garbage as compared to what they could be.
They both had the opportunity to step up and actually BE that "metaverse" that everyone was talking about and never took advantage.
Roblox chose "let's just let kids do whatever, stack money, and if there's a bunch of garbage that comes with that, so be it."
Minecraft chose extreme fragmentation and mediocrity. It's a wonder that MS hasn't run it into the ground completely, but also it's a wonder that they've done such a crap job with such a great opportunity.
I am certain there are very large regional variations on this. In the part of the US where I am, Android is considered the "normal" thing. Apple devices are around, but a very small percentage of the population and there appears to be little or no social pressure to prefer them.
That was the "vibe" of early mass computing. Anyone could be involved and anyone could make things better. It was fundamentally optimistic - felt a bit like Startrek NG :) Have a look at "Computer Lib / Dream Machines" by Nelson.
> the fundamental shift that transformed personal computing ... away from empowerment is the realization from industry giants that the platforms that they built ... they could exploit.
There has always a portion of the field that was looking to profit, but it wasn't the completely overwhelming portion that it has become, and the feel was for reasonable profit, not VC yields and "the company is the product" that came to characterise the field :(
Was it VC that killed computing?
The original 16K Apple II cost the equivalent of more than $5k. If you wanted 48k - and you did for any kind of non-trivial programming - the equivalent price would be $11k.
Computing isn't dead, it's just entirely commoditised. Most people who drive cars don't care about what's under the hood. They don't care they could build themselves a custom engine which is four times as powerful. They just want a transport appliance that works reliably and doesn't cost too much to run.
Modern phone-based computing is actually an insane success. Go to a restaurant and you'll get a QR code link to today's menu, which can be changed daily or even hourly. Sometimes you can even order from the table.
Go to a show, book a flight, or maybe just use some public transport, and there's a good chance you'll be able to use a digital ticket. Or at least use your phone for payment.
You can book medical appointments, fill prescriptions, track your fitness, and so on - all with one device.
Never mind basics like location and navigation, which aren't even surprising any more. If you ever use a paper street atlas you'll see how convenient they are.
The fact that you can't write your own code is of zero interest to most users. They don't want to write their own code. They just want something that makes life simpler. And mobile computing does.
The real problem are the social media monopolies and ad tech. But those are a political problem, not a technical one - specifically regulatory capture which neutered the anti-trust laws and allowed most commodity computing to be owned by a handful of monopolies.
I feel no real need to join the platforms tbh. Yeah, the platforms are easier, but that's to be expected, addictive fast-food is also easier. "Slow computing" takes some effort, and is not for everyone. But it's there for anyone seeking it out.
I'm on Matrix, run NextCloud, use Podverse for podcasts. Anyone can do that (well maybe NextCloud is bit on the technical side).
There was a brief moment ten years ago when XMPP ruled the world. Matrix sucks.
You're buying flexibility and freedom in exchange for complexity and service degradation.
If you think that stuff is worth it, you have to conclude it's like vegetables: they're healthier, but no one will want to eat them, even the people who realize they're heathier. :p
I come from a culture of: Boil you veggies to death, squash with potatoes, eat (most taste comes from the meat presented on the side). This can be delicious if the veggie is kale, the potatoes are squashed with butter, you add bacon baked in wine vinegar and honey, and eat it with some mustard. But generally the taste is boiled away. But, if you take your chicory for example, fry it, then add some orange juice, steam it a bit, make it into a "tarte tartin" with goat cheese, rocket and sun dried tomatoes on top... That's a new world.
Well, all the powerful entities in IT are.
Most people are on Android, that is technically Linux but not open by any standard. Most others are on iOS, that is even less open, besides technically being BSD.
Most of the ones trying to get an open platform to do something useful are stuck on Windows, and completely under the whims of a few powerful players.
That certainly played a role. But as your first citation already suggests, another major factor is that personal computers have steadily evolved into incredibly complicated systems, the details of which are way beyond what can be harnessed by anyone with the means and willingness to learn...
It's similar to cars. Nobody can fix todays cars besides professionals and I bet most have really no idea what's going. But most also don't care because they do not view cars as a hobby but as a tool that just has to work (it usually just works). If they are not required to fix them themselves they won't engage with it at all.
In computing the effect is to create a two-tier caste system where tech-savvy people have freedom and privacy and regular folk do not.
This is because while all software is expensive (in time or money) it becomes exponentially expensive as one tries to make it more and more usable by regular people. In other words polishing software is often many times more costly than creating the core systems that underlie it.
This is different from many other products. Imagine what cars might be like if the dashboard controls and interior finishings cost as much as 100X more than the engine and drive train. That's software. For most products the working parts are the most expensive part. For software the polish is the most expensive part.
When they became traps for our minds rather than tools. When business interested in attention, retention, churn, and keeping people paying started dominating the industry. When they became closed appliances instead of projects to build. When the Web made us depend on them.
I think the current state is mainly the inevitable outcome of most people wanting their computers toaster-like, i.e. appliances; and companies just doing their thing, which is maximizing profits.
Having lived through a lot of changes I think we're in a much better place than 20 years ago. The stuff I can do with my laptop to work exactly as I want it, while still being able to say access my bank account, is amazing.
In other words, I strongly believe that the "personal" part of the equation is strictly in the hands of the _person_.
I'm trying to figure out why these last 10 years or so I'm starting to feel like a luddite. I just find myself using tech less and less in my day to day life, for the most part.
There are certain things that are "new" that I wouldn't give up: grocery delivery services, online shopping and the ability to work remotely from home. In those specific areas, I can say that I "use" tech more.
But when I'm not working or avoiding driving to a store? I find that I have little to no interest in most of what modern tech has to offer these days.
For me, personally, something feels like it has been lost.
The closest analogue that I can think of is cars. I've never been a car person, but I live in the Detroit area and so I'm surrounded by people who are very much "car people." They hate modern cars. Why? Because to them, a car is not just an appliance or a way to get from one place to another. It was something that they took care of, maintained, nurtured, poured themselves into. And these days? Their local mechanic who they've known since they were a little kid can't even perform maintenance on it anymore. It needs to be sent to technicians who plug in a computer and will ultimately send it back to the manufacturer for parts replacements. Is this worse than before? Arguably it's better in many ways... probably safer. But it took something away from the people who loved the independence of being able to connect to a machine on an intimate level.
For me it was the "mobile revolution" that made me feel like I was becoming an old fashioned "car person." I never liked phones. I'm an asocial loner who doesn't like being interrupted by noise making devices. I don't want to have to remember to bring the thing with me. I never trusted the security and privacy of the things. I never enjoyed using a device with such a small screen and no keyboard (let alone the lack of a mechanical one to provide me with tactile feedback). So I never jumped on the bandwagon. To this day, I use GrapheneOS and a few FOSS apps (mp3 player and Signal to talk to my wife and that's about it), and I always have notifications turned off because I despise push notifications as a concept.
As a programmer, I have always felt this intimate connection between myself and "the machine." Although I never did a lot of low-level systems programming, that's where I find my heart going to these days. I feel so removed from the bare metal, like the connection has been severed and I've lost a lifelong partner.
But this still doesn't explain everything. It explains why I feel a loss as a PROGRAMMER and HACKER. But it doesn't explain why, as an END USER, I don't find myself using as many products and tech-related tools as I used to.
I'm still searching for those answers. Maybe it's me that changed, and not technology. I don't know.
If this wasn't the case, then locking down the user to one platform, as the article describes, would never have been so profitable -- most people would prefer customizability and "empowerment" over avoiding the slight inconvenience of using a different platform, and you couldn't build whole companies around locking down people at scale. But not only can you do that, those companies are hailed as being on the bleeding edge of innovation.
Now I personally find all this kinda depressing, but I think there's no point trying to turn away from it. Most people are simply very unlike the median HN user -- not in the sense of technical affinity or whatever, but in the sense of being curious and inclined to tinker -- and that is, alas, yet one more untoward fact to incorporate into your worldview.
What is the current opportunity for them? Nearly nothing. They might learn one system or another that can empower them. But the rest of computing is still like Philip K Dick's semi-proverbial Black Iron Prison. Maybe they can get off the wall in Plato's Cave for an hour & see a nice little tour around, but they'll spend the rest of their adult lives immiserated & chained back on the wall regardless.
I think people are very hungry for a better outlook with computing & e-social (which are so intertwined). Subconsciously they know the effort isn't worth the squeeze. Rejecting the premise of empowerment is actually just a value judgement, and a somewhat shrewd one.
The technics have to change first. The hobbyists, governments, research institutes need to drive forward some new malleable systems that actually get some inertia & growth & promise behind them. We need some users who spend the majority of their time not-chained up, who happily do most of their computing in the free. Computing needs a solid place to go, a worthwhile form of existence, before we can ask to attract people.
I stand with the banner mission of Malleable Systems Collective, a quote by J. C. R. Licklider, from Some Reflections on Early History (1986):
> The user wants open software, software that can be modified, and that can participate in a progressive improvement process.
But right now users have no where to turn to. This experience is unattainable, except on very small scales, and only after immense & painful toil. It's too early to tell what users want, because right now, there's hardly any choice. What choice there is comes with extreme sacrifice, is either disconnected & isolated alternative apps that offer little relative advantage & no interconnectivity, or difficult to host online services requiring dedication & committed-resources to try (YunoHost being perhaps the shiningest meta-attempt we could suggest https://yunohost.org/).
I dont have a great idea what this future looks like. (I have a bunch of personal possibilities I work on but they're just goes at what might be, not strong beliefs.) But we've chasing where the puck was, and it's no longer the objective.
> I also believe another important step to user empowerment is making programming easier for casual users, and making it easier for casual users to contribute to open-source software
In the history of Computer Science, it has never been easier to start programming than it is now. You have excellent support of WSL, access to Raspberry Pis, and $5 digital ocean cloud server to experiment with vast variety of technologies.
Lastly, I think there is some selection bias going on or looking at the past from the rose tinted glasses. Before computers were democratized, only people who were passionate about them used them. It's no surprise that they did all sorts of innovative things with them. Now that computers are widely available, the blame is being put on platforms instead of accepting the fact that the majority of population does not care about programming or computer science.
Have you tried moving the taskbar, customising the UI fonts/colours/sizes, or anything else in
https://en.wikipedia.org/wiki/List_of_features_removed_in_Wi...
https://en.wikipedia.org/wiki/List_of_features_removed_in_Wi...
https://en.wikipedia.org/wiki/List_of_features_removed_in_Wi...
> "In the history of Computer Science, it has never been easier to start programming than it is now."
Sure it has, 8-bit computers which started into BASIC interpreters in an instant, Visual Basic and Delphi drag-and-drop, HyperCard style environment - having to install WSL or buy and learn to access a cloud server is much more up front effort.
It is quite telling how far stagnation has gone when the latest and greatest operating system can basically do the same stuff as one from 13 years ago.
In the end of the 90s on my Pentium machine I had Windows 98 with 10 mbit LAN, web browsing, instant messages, code editor, word processor, spreadsheet application, video player, music player, image processing, content streaming, 3D accelerated games etc (on slower hardware).
The two major differences since then I can think of is that increased quality of media (images, videos, music etc) and interoperability, it is usually easier today to share documents and the like between different platforms.
Sure, there has also been improvements with system security and system stability, but neither of them adds to user workflow and productivity.
I tried Windows 11, but I had to go back to Windows 10, because I couldn't keep my old workflow, a workflow that has existed since Windows 95.
My own first computer that I bought for my own money, a Commodore 64, on that you just powered it on and started coding in BASIC.
So .. They dumbed-down the filesystem so they wouldn't have to explain it any more. And then they decided to never fix Finder/Explorer ever again .. and then they tried to get rid of it altogether. And even still today there are people who cannot understand the difference between a file and a folder, nor where or how they should organize their personal data in a way that makes sense. OS vendors decided to exploit this stupidity through anti-patterns that gave their bosses (the Marketing People) access to millions of users data...
The OS vendors dumbed down the user interface to make it more appealing to look at every day for hours and hours .. and then they forced us to watch ads in that interface.
They forgot that computers are general purpose devices, and tried to make us only do a few things with them. They removed the compiler.
They removed the compiler. <-- this is the most significant point at which personal computing went wrong. Instead of building a better compiler and a nicer machine to use, the OS vendors decided to make money off their developers, instead.
OS vendors stopped vending operating systems and became Marketing People™,®,© instead ..
Meanwhile, users get away with still not knowing what a filesystem is .. or why it matters to have a copy of that data out in the cloud, or locally on a machine in a place they can easily find it. The fact that this is still such a huge conundrum means that there really aren't any good computer operating system people out there any more.
They all went off to work for a FAANG and promptly retired.
I don't think so.
OS Vendors simply forgot that they service end users, and instead decided to service their marketing overlords. End users put up with it, because what choice do they have?
And so the actual goal is not to provide people with the best things, experiences, whatever. The goal is to make as much money as possible from them; the 'thing' being sold is just a hook.
You can see it all around; Razors and all kinds of other 'subscription' services, 'smart' devices, which could provide some kind of local smart interface that would truly give users more control, but instead insist on connecting back to a manufacturer's limited service, so data can be mined, adverts sold, etc.
The phone I bought four years ago has a bunch of great features now that were not present when I bought it, and I got those for free. Today, software engineers will go to work and write code to make it even better.
The subscription media services I pay for release more new content than I can possiblh watch or read or listen to, and I don't have to drive somewhere and pay for physical media.
And in winter months when it's snow and ice outside, I can start my car and get the heater and defroster running from my phone, even if it's in a garage a half mile away.
There are some aspects of some business models to dislike, but it is needlessly negative to complain about those without at least being aware of some really great changes since the days of "you bought this brick, it will be the same brick tomorrow".
And of course some subscription services are just that, you're paying for their ongoing work.
To be clear I'm not at all against companies updating their software, in fact they should make it all open so anyone can update it, or at least open up the hardware so anyone can make their own programs for it.
I think what you'll find with some of your free updates though, is that eventually they stop, when the product is no longer making money, or for example the demographic that is still using old devices is not one that is deemed profitable. And then you'll be all but forced to buy another device. One that has the same features that are now essential to the way you live your life.
In the hobbyist days before I got involved, you might build your own computer. The Altair 8800 and Heathkit type systems were meant to be assembled by the end-user. Of course, the tinkering did not end there, as they were programmable, and you typically didn't purchase a library of apps to load onto it, you wrote your own.
Now my family purchased a VIC-20 and soon after, a Commodore 64. Since I was a child, there was a heavy dose of gaming done on these machines, and I had joystick controllers and a library of commercial games that I could load. But, it was also very much an intro to systems architecture for me, where I learned how the memory is laid out, the function of various chips on the board, and how to program various subsystems from BASIC to do what I wanted. So at that point, the computer was really a good tool, and educational, with all sorts of ways that it could do my bidding.
In this era, BASIC and other end-user languages such as LOGO taught us how to program from the metal on up. And this was the joy of personal computing to us, the hobbyist.
Soon I graduated from high school and was presented with a real business-oriented 286 system, while my father got a similar one. We would run DOS and Windows 3.1. There was a lot less of programming it ourselves and a lot more of loading commercial software on those. So I think that was the turning point of home computing, from hobbyist workbench into a business appliance that was a real means to an end.
My later computers, a 386 and a 486, were partly gaming machines and partly Internet terminals, and of course my phase of programming it myself had fallen by the wayside, in favor of loading ever more complex applications and operating systems, and that was where my tinkering focused for the next phase: the administration of the OS. So in my career I became a systems admin who enjoyed configuring stuff, taking security measures, and tweaking the OS for performance. I also enjoyed building my own computers, as they call it, "LEGO for Adults".
A year and a half ago, I shut down my Linux system. It was the last desktop system that I owned. I had assembled it myself from discrete parts. I had tweaked the Ubuntu OS endlessly. I no longer needed this power-user stuff.
Today I get by on notebook computers only. I run Windows and ChromeOS. I'm truly an end-user, if on the power-user side, but I never program the system myself anymore, and I don't tweak the OS like a systems admin, I just dutifully install the updates. Because my computers are "in production". They serve me to work for my employer and to manage my household. They are an integral part of it, rather than a hobbyist's workbench. I don't know if that has somehow "gone wrong" along the way, but it has definitely changed character since 10-year-old me picked up that VIC-20.
From self-actualized and in control to a complacent slave at the mercy of your corporate jailors
My trajectory has been the opposite, going from not knowing anything (because I didn't need to, as a kid using Windows 7) to building my computer and then, a decade later, using Arch btw (because I wanted to). That it might one day reverse terrifies me, given the uniquely rich dystopian potential technology seems to have as it evolves.
I have an EliteBook running Fedora, and some virtual private servers to run Matrix and NextCloud. Fedora was solid for about a year, and then it catastrophically crashed. I had backups so I could get my files back, which was a good thing, since NextCloud's encryption relies on an unencrypted local copy (in my particular version). Decrypting files stored on the server was onerous, at least in the particular combination of server/client I was running. Since I rely on those files (OCR scans of letters, building blueprints, photos and a bunch of other stuff) I found this unacceptable.
For the phone I have a Fairphone 4 running Murena's /e/OS (which is a stupid name, by the way) which works. GPS was crappy though, and aside from messaging that was all I used my phone for.
Yes, it is important to have some measure of freedom, and I am still vigilant against locking myself into systems I'd have a hard time getting out of. But at some point pragmatism wins. I never wanted to tinker (on production systems, anyway), I want stuff to work so I can go about my business.
So now it's an M1 Pro MBP and the iPhone 8 I had in a drawer. I set up some automation to copy Photos out of the library into flat JPEGs and I copy the iCloud data to an external drive (in flat, open format) regularly. Other than that I rely on Apple to keep my OS secure and hardware running.
Some part of me misses the feeling of being 100% in control. On the other hand, this just works, and I have little issue throwing money at the problem.
Between idealism and pragmatism, the latter definitely wins when push comes to shove.
It's also easy for me to posture as though I use GNU/Linux out of bitter, noble sacrifice and righteous fury toward Microsoft, but that's like 1% of it. The 99% of it is pragmatism. I happened to have lucked into a set of priorities where Linux is the more pragmatic choice (as long as you pay the upfront cost to learn it). I don't need to use PhotoShop or audio synthesis software, for example. I just program. So, for me, Linux works great! :D I also play (mostly non-AAA) video games. If Valve hadn't supported Linux, you best believe I wouldn't have survived on it.
So, actually, my real fear isn't that I get complacent about my ideals; rather, it's that my priority matrix never shift such that the optimum solution begins to run against my ideals. (That's partly the case now that Microsoft owns GitHub, since I use GitHub, and I try to forget that Microsoft bought it every time I use it.)
> Today I get by on notebook computers only. I run ... ChromeOS.
I think ChromeOS is the diamond in the rough. Updates are a breeze, it Just Works and one can easily run Linux via a well-integrated virtual machine if they want.
These sweeping but vague explanations are tempting, but I think the cause is actually a lot more technical and random: the OS design culture that spawned Windows, UNIX and macOS pre-dated the internet. The internet was a fundamental shift that changed everything, and yet the NT and Apple design teams all traced their roots back to the pre-internet era (and Linux just made a UNIX for the PC, same thing, design rooted in the 80s). Their whole OS design thinking was fundamentally pre-internet and never caught up. Even basic tasks like uploading files to a server through a GUI STILL aren't well supported in desktop operating systems, in 2023!
What platform did fundamentally get the internet? The web - born of the internet, made of the internet. Web apps don't even start up without the internet. Whilst Microsoft snoozed, Apple thrashed around and the UNIX people spent all their efforts on cloning the past, web browsers became the new way software got distributed. The web had so many benefits over writing PC/Mac apps. Suddenly you could iterate every day, you could write the bulk of your software in nice high level scripting languages, your software was accessible by just typing in its name, you could use databases instead of designing binary file formats, app UI was automatically portable between operating systems, typography and layout were taken seriously, people could easily discover and access your software with one click on a link, the platform had a nice open easy to use protocol set up for you out of the box. And dozens more.
So many wins for developers and users in there. Unfortunately, it was never meant for this. People spent vast effort hacking the web into a barely-passable app platform because it was the closest thing to an Internet OS, and nobody else was offering that. But the web was simply never thought about by its developers in terms of empowering personal computing users. Whilst the people at PARC, NeXT, Apple, Microsoft and so on spent lots of time thinking big 1990s era thoughts about making programming easy, computers powerful, component systems, filing systems with search, AppleScripting, better window management and so on, this utopian way of thinking never entered the culture of browser developers.
There was a brief attempt at some sort of structured higher level planning, that open hackable spirit during the W3C era, but the XML stack was made by academics and despite good intentions a failure. So browser makers just kept randomly flinging stuff into HTML, but without the animating forward-looking spirit of the PC era, in fact without much of a vision at all beyond trying to keep the show on the road and better competing with 'real' app platforms.
It could easily have been very different. If the people doing OS development in the 90s had fully understood the importance of being an internet OS we might have had things like a global filesystem namespace, start menus that could directly launch sandboxed OS-native apps from a domain name, internet sharable clipboards, who knows what else. Unfortunately that's not how it worked out. Today OS research is basically dead. ChromeOS is the newest and probably the least powerful, least ambitious OS ever made. On the rare occasions people get to make a new OS they never make it past micro-kernels. The kind of sweeping ambition that launched the PC era is gone.
Both the rise of platforms & the smartphone are tightly interwoven around a new connected axis of computing emerging & being useful. McThrow somewhat generously includes the web in his "personal computing" vision but alas, even as a huge web fan, it's pretty clear that the brief "intertwingularizing" moment (which was extremely pro user) was giving way to Service as a Software Substitute (SaaSS), to a de-personalized experience. The relevance of desktop faded, and relevance faded because users were well served by new connected services. Software effervescing, sublimating up into the cloud made it easier to share, made it easier to access to demand, freed us from worrying about backups.
I think this is one of the critical junctures that Linux and the rest of the Desktops has failed to internalize, as a new need: the need to be connected. The need to be more than just isolated personal, the need to be available, the need to be connectable. I don't know quite what this future looks like or what it implies, but the "Linux on the desktop" spoke of competition with an outmoded, fading mode of computing. What was really needed was a further push onward, was a backbone for inter-app systems & development.
One of the few efforts here is one people love to hate on (even though I think most have a superficial at best grasp of the technics): One Laptop Per Child, OLPC/XOPC. My word what a wonderful & interesting system, powered by a grand already-well-adopted-on-FreeDesktop/Linux desktop bus (DBus) that underpins a wide variety of the software matter about. And then the team went further & built DBus Telepathy[2], to connect DBus instances to one a other, across the local 802.11s mesh the laptop had, to create a very compelling shared multi-player object stage that apps could use to create shared experiences. Stunning smart extension of existing highly personal intra-desktop tech into a connected form, a connected media; simply brilliant.
There's a couple reasons for hope in my view. First, the connected age is new & we are iterating & learning what broader connectivities work, and in there I think is the real base of power, the higher path that will be compelling for folks to get into. Nostr in particular has a very grand set of Nostr Implementation Possibilities[3] that shows how flexible the simple protocol is. But works like Dat, Hyper, ActivityPub, AtProto/BlueSky, Solid show a hunger & willingness to find out, to explore, at the base of a big tree of possibilities that more personal yet connected futures might emerge upon.
I'm also happy to see computing operations getting mature. It was a free for all, learn everything do everything, everything is different & unique, there are no standard all-encompassing routes until the last half decade. Love it or hate it but there is nothing as integrative & consistent for operating computing resources as Kubernetes. It's an incredibly simple pattern at heart, applicable to almost all computing, and its aspirations of being autonomic, self-maintaining, are such a compelling human-liberating spirit that server-hugging desperately needed. Even if it's not Kubernetes in the end, the spirit of an integrating platform for computing that can reasonably self-manage is a must, if we want personal computing to win. Personal computing has to become more than a machine, it just has to, the demands are too high, and we are not going back, and thankfully, we now have good mature sensible ways to weave together the thousands of different threads of running these online endpoints ourselves. We'll keep getting better with smarter packaged implementations, and better tooling; the momentum is for real on & it's so good.
The hardest to hope for win is the Web, but it's to most essential. The web has a lot of good-hearted people who care about personal computing. But the forces stacked up are huge; the industrialization of the field has been epic. (And getting ghastly worse, with WebGPU canvases threatening to turn the screen into just a bunch of pixels, see Hixie's Towards a modern Web Stack[4] and Flutter CanvasKit implementation). Alas in my view many of the good, spirited radicals differentiate themselves negatively, regressively, with back to the land ethos rather than having willingness or interest in forward advancement, when what we need is ambition & striving & exploration forward, particularly by rejecting industrialized best practices & asking ourselves again & again what best serves the user, what is the most open & empowering routes we can find? What course forward would we lead?
I think there are some HyperCard-ish futures about in easy-ish reach, some explainable observable quick & interactive modes of computing, if we go hunting, and that the web is already almost the perfect material/medium. WebComponents are a key abstraction, and I'm not ashamed to say how thankful I am that the biggest most powerful tech companies are quickly adopting it en-mass & seeing great success (YouTube, GitHub, Google, Salesforce[5]), and that although underpublicized, it creates learnable moments & good offshoot tech (Stencil.js, Github's Catalyst, Google's Lit). Embedding app state into the DOM makes it malleable, makes it learnable in a way code isn't. My hope is we can start to create more visual editor like systems, where the page itself exposes it's own tools to better show the user the objects on the screen, and let's them play and explore more. Make the medium more live & interactable.
Yet this is still not enough. A more user-friendly web page medium still, and better ability to host endpoints, still leaves a whole question of what's in the middle & how does it work. It returns to the protocols like Nostr I mentioned above. Figuring out how to personalize the protocols, how to Augment Intellect (the highest mission, the Engelbartian vision) in computing needs so many levels of the stack to be illuminated & graspable.
And above all that stack, what is the experience? The world of applications, what most users use, is a lot of isolated systems. This hearkens back to the point of this submission, that personal computing had fallen, and in many ways it's because the at hand value to capture was upmarket, was higher level concerns: apps. Personal Computing fell away because each app is it's own fief, it's own walled garden, and we didn't emerge a broader system of computing that meaningfully intersected with and connected with thr apps. Isolated industrial experiences has been the name of the game for decades now, that brief Intertwingularity/Cluetrain Manifesto zeitgeist the late aughts aside. I've talked a lot about stack & technics, but ultimately a more integrative holistic stack has to be here to be useful, to serve the needs we have. Making HyperCard-y experiences that can interlink & coexist together on the canvas feels like a maybe possible vision here, something works like OpenSocial[6] were trying for during the greatest headiest API-based moments of computing history. It was a greater effort, but that attempt to stitch together widgets & data has some notable overlap with above mentioned Solid.
Overall, I'd emphasize that for computing to be personal, it has to be personalizable. It has to be the users, in some degree. On these simple grounds, I can think of no more direct & overarching a cause than the Malleable Systems Collective[7], which is a wonderful home-base for exploring what computing do, how it might get out of the imprisonment of the data keep, oops i mean data center, & back into the hands of users.
[1] https://www.chrbutler.com/what-i-want-from-the-internet https://news.ycombinator.com/item?id=35852516
[2] https://aosabook.org/en/v1/telepathy.html
[3] https://github.com/nostr-protocol/nips
[4] https://docs.google.com/document/d/1peUSMsvFGvqD5yKh3GprskLC... https://news.ycombinator.com/item?id=34612696
[5] https://levelup.gitconnected.com/web-components-at-big-tech-...
To really do it right, you have to open up all of the stack starting down in the hardware, and then socially engineer people to less complex and bloated standards. Without losing valued functionality. A very tall order.