ocathain.ie/articles/for-want-of-a-more-limited-computer.org

#+TITLE: For Want Of A More Limited Computer
#+DATE: <2023-01-26 Thu>
#+DESCRIPTION: A lament against certain design trends in computing that has caused the apparent loss in a few of the things I found fascinating about computers in the first place. Or, in other words, a verbose "Damn it, I missed the cool stuff!"
Modern computing is truly a marvel. I have a device sitting in front of me that
at my beck and call can send a request halfway around the world for 2 million
bytes to come screaming down a complex system of connected computers, which is
then decoded and displayed on a monitor refreshing one hundred and forty-four
times a second, all taking less than the time it takes me to blink. All of these
computers working in unison, to deliver to some random guy a distorted image of
the actor Bob Odenkirk with the caption: "Your Honor, do you expect the jury to
believe that a shrimp fried that rice?"

/What the actual hell?/ Sixty years ago, the idea of a /personal/ computer that
could fit in a house was pure science fiction! Then when /microcomputers/ came
along, it was still preposterous to imagine the average consumer could afford to
have a computer, much less access to a worldwide network with "free" (more on
that, later) services and information which can be summoned at a moment's
notice. And then you look at today's reality, and this amazing technology is
/real/ and /available/! So what have we all done with it?

* Part One: One Man /Can't/ Know It All.
I suppose I better start with what I've done.

I've been fascinated with computers ever since my family's first computer, which
sadly I do not recall the model of. It was a generic enough Windows XP machine,
that's all I can remember. But it was more than enough to get me hooked. Here was
a machine that could do many, /many/, things, and my child mind was determined to 
know /everything/ about it. Many a night was spent poring over menu options,
trying to find every little thing I could learn about. Broke the thing, a couple
of times.

I pored over that machine, hoping that I'd be able to understand every little
thing about it. Granted, I wasn't suddenly going to grasp the intricacies of
networking, or operating system design, or hardware architecture, especially not
by just clicking buttons in a Windows XP install. But it did leave me with my
main hyperfixation: All things computers.

I began to program when I was in third class, when my teacher at the time
recommended CoderDojo to my parents. I played around with some Scratch, which
was fun, plugging together little games, I did a bit of HTML and CSS, which
wasn't really for me. I wanted to know *more*, though.

I started to read a bit about programs I used and games I played. And one day, I
found something really interesting: The /source code/ of Quake III Arena. It was
my first time looking through a codebase of that scale. I glanced through it,
until I came across quite a famous function. Fast Inverse Square Root. I was
utterly mystified. This was the kind of thing I was enamoured with. A solution
to a problem that was a mystery itself. "0x5f3759df"? "Evil Floating Point bit
level hacking"? It was from there I knew I wanted to work with computers, in any
capacity possible. (It also started my love of arena shooters, because who looks
at the source code of a game without playing it?) It was also there that I began
to learn about Free Software, as the GPL was what allowed me to have my little
revelation. I saw how people could collaborate, improve on each other's work,
and create something together. I steeled myself, sat down, looked up some
programming tutorials, and dove in.

At this point I made a pretty big mistake: I started bouncing around languages,
depending on whatever mood took me. I never got any further than implementing
something trivial for a few years. I still regret that. I should have realized
that I wasn't really learning anything new, or developing any useful skills. I
finally dug in, and picked C to learn "properly", with the goal that it would
help me understand "what's /really/ going on." 

It did that, somewhat. I tried to learn some x86-64 assembly, to get an even
deeper look, and was scratching my head at the sheer size of the instruction
set. But then I finally made my realization: modern personal computers are far
too advanced for one person to understand, fully. You can understand the general
concepts, even know a few "secret sauce" tricks, but at the end of the day, you
can't truly understand everything. Not for lack of trying: even if you were a
genii, you'd have to work at numerous companies to gain access to the documents
and designs you'd need to do so. Bit too much of a time-sink, even for the most
dedicated individuals, for what is ultimately a child-like goal. I could try to
develop my own operating systems, but the GPU that's sitting in my PCI-e slot is
far beyond my capabilities to work with. I wouldn't ever understand every
instruction that my CPU could run, nor am I meant to. No matter what I wrote, I
could never be sure that I was in full control.

Granted, I could do a project like jdh's computer and graphics card from
scratch, but that's an altogether different goal. (That can be found on
YouTube) Designing a computer isn't the same as learning everything about a
computer you have, though it is perhaps just as or even more impressive. It was
somewhat possible in the days of 8-bit home computers to understand everything
about your computer, not easy, but possible. At the end of the day, I have to
concede to trying to understand quite a bit about my computer, which just
doesn't have the same /ring/ to it. 

Sometimes I wish I did grow up in the time of the 8-bit home computers, and to
be able to have my child-like wonder satiated by manuals, schematics, and
software, to explore and squeeze every last bit of performance out of one of
those little machines. To this day I'm still fascinated the demoscenes of
devices like the Commodore 64, or the ZX Spectrum, and have often considered
trying my hand at them, but ultimately, I've just never gotten around to it.

I hope that in the future, both libre hardware and software will be able to
create a new useful computer that I /can/ fully understand, but I'll have to wait
and see on that one. It's simply a fact in our complex, modern computers, that
there's too much for one man to know.

Then again, maybe it's me that's not smart enough. 

* Part Two: The Internet /Was/ A Big Place.
For almost as long as I've used computers, I've had some form of Internet
access. From a USB mobile broadband modem, to a mobile broadband hotspot, to a
terrible copper-wire router with an attitude problem and a tiny data cap, to
a finally tolerable router with unlimited data. My earliest memories were of
playing random flash games, and very little else. Had I found the right
resources when I was younger, I may have been much farther along in my
programming journey by now, but that would have been difficult enough.
I wasn't too much of an explorer in the early days of the internet, holding
myself back due to tall tales of the dangers of the web, keeping under the
limited data caps, and most of all, avoiding the anger of my parents, afraid
that I'd provoke anger by discovering... something. I wasn't sure what at that
age, but I played it safe. My parents were right, of course. There was a lot of
stuff I wouldn't have understood at that age and likely would have been harmful.

Had I been older at the time, I would have been able to experience the death
of the Web 1.0, and the transition of content to social media. I personally
think that this is what made so much of what toxic, harmful design we see
today possible. But more than that, this made the internet feel a /whole/ lot
smaller. No longer are you able to stumble across niche communities by chance,
not with search engines pointed squarely at the landfill of garbage and
mind-numbing content perpetuated by whatever platform was "in" at the time.

And that's what I think is sad about the modern internet. Yes, you can find
cool sites from time to time, but with everyone's eyes glued squarely to
hostile sites designed to grab their attention for longer and longer, these
cool sites are cast to the wayside in exchange for the next generic blob of
popular "culture." I wouldn't feel so annoyed at this if it were the good
content on these platforms rising to the top, like the layer of foam on a
soft drink, but it's often the most mind-numbing, low-effort,
just-entertaining-enough tripe that ends up the algorithms fancy for the
increasingly shortening attention span of the users.

And it's this plainly *hostile* design that makes me sick to my stomach. These
platforms are so clearly designed to create addiction in their users, through
slimey techniques like infinite scrolling and push notifications, where systems
are built solely for the purpose of maximising the time a user spends on the
platform, through the mining of data of millions and millions.

I'm not going to pretend I'm a veteran of the older internet and that I know
exactly where it all went wrong, nor do I intend to gatekeep something I have
no right to gatekeep. I will, however, say that making content easy to publish
on the Internet, albeit inside a walled garden, seems like a great way to
allow creative people to do their thing and show it to an audience, but it
also allowed people who just wanted to grab a brief moment of fame to trumpet
in and throw whatever at an audience to see what stuck to do so. It created a
culture of fleeting attempts at striking it big quick, which leads to a deluge
of garbage flooding the public conciousness.

But this is just a secondary effect, this dilution of quality and
creativity. The primary effect of this transition was taking a decentralized
system and centralizing it into hubs, where a singular corporation would have
jurisdiction. *This* makes me angry. The freedom of information that is the
promise of the internet isn't possible with a single party in control of said
information. Yes, it's more possible than ever to host information outside of
these networks, but the network effect is a powerful thing. People don't hear
about things if it isn't on their various newsfeeds. It's created a partition
between an average user and one which is aware of communities outside of that
bubble. It's not the users fault, not at all. Once the right person had the
idea of creating that bubble, there wasn't much anyone could do to stop
them. Of course people were going to use something that made the hot new thing
easy for everyone. Of course buisnesses were going to flock to try and make
money in this new frontier. Of course it was going to become harder and harder
to find the good from the bad. I speak with hindsight, and I know it wouldn't
have seemed obvious from that point in time, but yet I still find myself
annoyed.

Yes, that's probably a me problem. I'll go touch some grass, now.

* Part Three: Holy Hell, It's Getting /Worse/?!
It's easy to look back at technology with rose-tinted goggles. Taking those off
for a moment, I remember my first computer being slow and unresponsive. I could
stare at the Windows XP logo long enough that it would leave a ghost image on my
eyes afterwards. I'm rather surprised that I managed to get anything done on
that computer at all, a younger me having much less patience. Thankfully, now in
2022 I have a reasonably powerful computer, so all my applications should run
quickly and snappily. 

Right?

/Not a hope in hell./ Somewhere along the line, computers became powerful enough
that optimization was not the be all and end all. Of course, I can't hold that
against anyone. Not everyone is going to be the super hacker that does
pico-optimizations to save a grand total of 3 CPU cycles in a massive
application like a CAD program or a web browser. But it seems more and more like
modern programs do everything BUT optimize themselves.

I think the biggest example of what I mean comes from Electron; a nice idea in
concept, by allowing web developers to create native applications with the
languages and design capabilities that they're used to. Certainly, it's easier
to theme and design than something like GTK, that's for sure. But it comes with
the trade-off of running an entire browser engine to use your application. At
that point, I'd rather just run your application in my actual browser. I think
the most egregious examples come from applications which require you to be
connected to the internet to use anyway: for example, chat programs. However,
the one that really took the cake for me was Balena Etcher; an image-writing
program that uses Electron to provide a pretty GUI. Not the worst goal, but to
run a browser to copy some bytes? /Really?/

And if you ask me, optimization is even more important nowadays. I don't want to
be wasting cycles to display an application that could be native when I'm on a
machine with a battery, whether that's a phone or a laptop. At least offer me
the option to use a native window toolkit or other solution, or if you're
running a service, to be able to use my own client.

But I'm sure it's just me that needs to get a better computer. Or I just need to
update for the 50th time today. I'm sure whatever "various bugfixes" have been
done will solve all of my problems! 

* Part Four: Privacy, or what's /left/ of it.
Privacy is something that's important to me. Do I have things to /hide?/ No more
than any other person. I don't want people to know things like my bank account
information, or the amount of times I listened to the /Wrath of Khan/ theme loop
in a single sitting. But what I don't understand is the reaction people give
when you tell them you'd prefer not to be tracked by gigantic corporations who
log everything you do on your devices and on the internet.

Suddenly, you're accused of being some kind of conspiracy theorist, or you get a
response along the lines of "But why would they want my data? Who cares?". The
simple fact of the matter is they do, and they use it specifically to try to
manipulate you, either through ads, or as we found out about companies like
Cambridge Analytica, manipulate your political opinions to match whatever the
highest bidder wants.

And so, as I began earlier, most "free" services you see on the internet aren't
free at all. It's a transaction. "Let us track you relentlessly in order to
learn as much as possible about you and the people you connect with, so we can
sell that data and manipulate you, and you can put funny pictures over your face
and send them to your friends." Sounds like a much worse deal now, doesn't it?

And really, it /is/ becoming worse and worse. With the advent of machine learning
and related technologies, who knows what fun new ways to track and analyze us
corporations are working on? How much can you manipulate a person by knowing how
their eyeballs move inside of a VR headset, for example? There are people who
likely have a highly paid job right now to figure that out.

As a rule of thumb, if you or that computer-savvy friend you have can't find the
source code somewhere for whatever service you're using, or the Privacy Policy
reads like /Finnegan's Wake/, you're going to have to do a lot of work to not be
tracked. If it's not critical to your life, I'd say drop it, or start
researching how to minimize the collection of data from that service. It's
incredibly fortunate that there are dedicated people who develop software to
help with things like that.

Or, I suppose, you could wear a tinfoil hat. I'm sure that'll be just as
effective. 
* Part Five: So what are you going to /do/ about it?
To be frank, I can't change all of this. This is the route that design has gone
down, and after a certain point you can't really fight against the tide. I will,
however, try not to worsen the problem. And at the end of the day, it's still
really cool, and I'm still going to try and learn as much as I can.

Just because you can't know it all doesn't mean you get to throw in the towel,
and so I'm not going to. I'll just write my programs, and hopefully, they'll be
useful to someone.

But all the while, I'll feel the want of a more limited computer. One that I can
understand, fully, top to bottom, back to front. One that is not limited in
usefulness, but in design sprawl. One that I can run software on that doesn't
trample over my privacy or pin my processor to 100 percent.

Or I could scrounge up a fortune and become one of those diehard /Amiga/ fans.