okay! we would like to start a discussion about what it would take to make Linux *fun* for newcomers. please chime in with your ideas!
we've personally always found it fun, we're in the crowd where it's more important to be able to tinker with things than for the things to actually *work*. however, we recognize that this is a minority position.
we do think that there needs to be a lot of work on usability. just, like... taking all the "how to debug" wisdom that's currently spread out across the ArchWiki and a million Bugzilla threads, and turning it into UI that guides people towards the solution.
we also think that there are significant areas of the FOSS ecosystem where nobody is taking responsibility for their overall health, and somebody needs to do that.
neither of those things is directly about FUN though.
in the 80s, we could get a desktop toy that made mysterious footprints across the screen of our Mac. while we were trying to do other things. for one reason and another, stuff like that got lost in the shuffle and isn't around anymore.
what other ways are there to make computers fun?
This is true, for sure. It's a shame it's such an obscure thing right now. Needs a tech writer's attention to make it easier to get into.

We kind of feel like if people want game consoles, they know where to find them, but... this deserves to be mentioned.
ooh! super cool! yeah, window manager customization is hugely empowering.
then there's this very interesting suggestion to *remove* the GUI

yeah. this seems really important, being able to roll back to a known-good state.
ooh. this is such a good suggestion!
re: UI customization again. absolutely agreed. we get *why* it's rarer today - it makes it easy to break things, and corporate OS vendors would rather you see their brand elements than your own customizations - but it's deeply saddening.

trust NireBryce to have thoughts on how to make *terminals* more friendly. seriously, this is great to see, especially since it dovetails with one of our back-burner projects.

100%. there needs to be a lot of attention paid to making sure users are *aware* when they're customizing something in a way which makes the UI harder to make sense of.

this is such an important point. it's even something that exists, just... it doesn't have as much mindshare as it used to.

another vote for making it safer to experiment with internals. we're personally enthused about declarative-state distros such as NixOS and guix, but there's a lot of work that needs to be done to make them beginner-friendly.

and another vote for GUI customization.
hmmm, yeah... we do agree with this. even when we were beginners, wizards rubbed us the wrong way because we didn't learn anything from going through them. it just does things for you, it doesn't tell *you* how to do it.

some of the earliest consumer OSes came with tutorials on how to do basic tasks, and the way those tutorials were designed - with basically fake desktop/CLI setups that looked real but had the sharp edges filed off - was a good approach.
for example, we had an MS-DOS one that explained how to work with files and run BASIC. we find that particularly notable today because it was an IBM thing and IBM is not known for aiming its products at young people... yet they did, that time.
we agree, strongly, with this train of thought about making the GUI configuration stuff be explicit about how it relates to the underlying, text-based configuration.

and then, of course, we have one friend who's already been working on building the future of computing, in VR. <3
don't miss the concept art for this one
more thoughts on open-source games designed for tinkering. yeah, this is the fundamental challenge... the kind of work we're describing can't be done in a for-profit context because empowering users is directly at odds with profit.

it took us a long time to understand that, and to believe it, and we don't want to clog up this thread with an analysis of the broken incentive structure... so we'll just note it in passing.
yes. we agree with this. it's much simpler from a user's perspective. config syntaxes are a well-understood problem space, we don't need a million different ones.
heh! yeah... now that we think about it, we know what window manager we use because we spent time messing with the low-level parts, but there isn't a straightforward way for a newcomer to find out.
yep. installers are important! it's the first experience people have with your distro. it needs to make a good impression.

here's a remark building on the point about GUI config that explains what it's doing behind-the-scenes. we agree - if the GUI is trying to make things simpler, it needs to be able to cope with weird situations!

a thought on how to implement rollback, which was mentioned up-thread (BTRFS is a filesystem with copy-on-write semantics)
yes, absolutely. this is a fundamental tension that applies to all of the ideas people have been bringing up. it's really important to have a principled approach to it, rooted in solid CS theory.
yeah, we've also put a lot of thought into whether something like HyperCard could make sense in the modern environment. we think it could, it just needs some rethinking.
here's a good example of one of those trade-offs between complexity and fun.

visually annotating workflows is a really interesting idea, very different from the other suggestions in this thread, and we like it a lot. most beginning computer users aren't even aware of workflows as thinking-tools, but everyone uses them.

oh wow yes. this has a lot of potential.
we're going to also link the second tweet from that subthread, so it doesn't get missed.
more thoughts on rollback. there's a design tension here, because corporate users want immutability for very different reasons than individuals do, but it still seems worth thinking about.
yes, this is incredibly important. even giant corporations seem to think it's acceptable to pinch pennies and ignore non-English-speaking users, but it's really, really not.
some thoughts on what to do instead of wizards; as noted up-thread, wizards don't really *teach* anything.
here's an entirely different angle on rollback. it's good to be reminded of this kind of approach. in some ways, the lower cost of storage has made things harder because it means there's more mutable state...
and another point on managing complexity
yes, 100%. bundling a development environment is huge. every 80s microcomputer did, and things really started to go wrong when development stopped being a thing everyone was encouraged to look into.

we would like to take a moment to say that we couldn't do threads like this without all our awesome mutuals. thanks so much, all of you!
some thoughts on how to navigate that complexity/usability tradeoff
another vote for UI customization. we do understand that what we're asking in this thread isn't something everyone will want, and that's fine!
we're prioritizing FUN not because it's the only possible goal, but because we have a larger social purpose here. knowledge really is power, to a far greater extent than most people appreciate.
neat to have creative ideas like this! we're cautious about this one because achievements systems are fun precisely because they turn a raw, unfiltered experience into a packaged one where you're told what to feel... but it bears thinking about, still.

ooh. we have NOT tried this. it sounds really cool! "playable archaeology" is an amazing phrase.
oh wow, Vim Adventures seems like a super cool idea.
responses seem to be slowing down, so we'll use this as a chance to talk a little more about why making computers FUN is so important to us, personally.
briefly: the tech industry, today, has become more and more about corporate, hierarchical power structures in which end users have very little control over their experiences, over their lives, over what's done with their data.
one of the threads that runs all through our life is the discovery that knowledge really is power, that as long as you keep the hacker spirit alive in your heart you are still free in an essential way no matter how oppressive the systems that try to control you.
the tech industry is in the middle of a phase transition, which other industries have been through before. in another twenty years, computers will no longer be the new, cool thing. they'll be something more like automobiles: nice to have, but primarily utilitarian.
we know this because it's what's happened to *every* industry which has held the societal role that computers hold today.
however... we have a choice about how it turns out.
our friend @beka_valentine likes to talk about the analogy to the early labor fights in the garment industry. the creative aspects of the work were compartmentalized off, and minimized.
@beka_valentine there's a certain, fuzzy sense in which tools like Kubernetes are analogous to factory looms: they are designed to serve needs that only exist for giant corporations, and individuals have no use for them and no clear place to start learning about all the complexity they add.
for anyone who grew up in an age when computers were meant to be tinkered with, first and foremost... if you want to preserve that culture, this is the moment to fight for it. these next few years will shape a lot of things down the line.
this is the moment to put our egos aside, to reject elitist attitudes that hold FOSS communities back, and to make it a point to uphold the best aspects of the hacker spirit, what computing COULD be.
it's fine that not everybody will want to tinker with their computer. it's incredibly important, though, to preserve a world where people CAN.
there are still people who tinker with cars, today, and that's really great. there needs to be more of that. even as computers go through this phase transition into being an accepted part of the world rather than something NEW, it is entirely possible to preserve what matters.
we've written before about how we see the possibilities that technology offers as a gift from the universe.

let's keep computing WILD, not tame.

enormous and fuzzy and full of possibilities.

thanks <3
people are still contributing ideas, so we'll keep linking stuff :) both technical and cultural ideas are in-scope!

yeah, this definitely seems relevant! everybody hated the paperclip in the day, but that was more about how intrusive it was...
regarding manual pages and how to make the documentation more explorable. we've had this thought before, too, we think it could help a lot.

yeah - showing people that they CAN do this stuff is important.
we honestly get really annoyed at corporations whose business model is "you will never be able to do this stuff, so we do it for you!"

because then there's a constituency who wants to keep things difficult!

yeah. we sympathize with this. a lot. it's partly a community-building problem; nobody is taking responsibility for creating a culture where things get explained...

this is an AMAZING idea. very difficult to implement, but it could be very powerful.
:D yes! by all means look at what kids who get computers actually do with them, for inspiration. kids are the best connoisseurs of fun.

more points around the "factory loom" thing.
for sure. being able to boot into a LiveCD and have a fully functioning environment with all the basic tools is pretty amazing.
knowledge bases are really important. the ArchWiki is by far the best one out there for general Linux topics; it's a shame that it's so closely tied to a community centered around elitism.
yeah, this is a huge problem and one of the core things that excites us about declarative-state distros such as NixOS and guix. we're less excited about Ansible because it's still mutable state under the hood.
for sure - making it easier to get to the documentation is a big deal.
very good point. gaming support doesn't just mean the games themselves.
more thoughts from watching what a kid wants

(disclosure: we deleted and re-made this tweet to fix threading)
good observation. yeah, maybe it's about knowing where to go to find things that will definitely work, more than it's about stripping away the things that don't always.
right, yeah! this is what we personally have with NixOS, but... we're experts. it's not something that beginners can really do, right now. it should be.
ResEdit was amazing. (sorry - this is a quote tweet of a quote tweet, you'll need to click through.) yes, there are different schools of thought about it, but it's worth at least understanding what the good parts were.
hmm... sure, okay, but our core thesis here is kinda that if you want to empower users you can't wall off any area of the infrastructure as "don't go here". still, it's important to keep the counterpoints in mind.

more thoughts about going even further than open-source games designed to be hacked
lots of input devices (we have a friend who wired up a big industrial "stop the machine!" button to invoke one o their cooldowns in an MMO)

maybe we don't need to move on from the desktop metaphor at all :)
yeah... so we've spend many years studying language design, and as part of that we've looked at visual programming environments, and very few of them achieved their goals. nonetheless, there's stuff to learn from them (and we're huge fans of HyperCard).

neat idea. this ties into our thoughts about the problem with wizards being that they don't teach you how to do it yourself.

oh wow. integrating chat into the environment so it's easy for people to find help and community. that's a really cool idea.
oh goddess, yes. tech writers can do SO MUCH to help any technology, and FOSS communities really under-value them.
very cool thoughts. some of this you can get with NixOS, but it requires users to already have lots of low-level knowledge.
and here's an amazing user story that definitely can't be done with any existing tech.
more thoughts on how to make the stripped-down "just the important parts" thing not disempower users.
yeah this actually ties into another of our back-burner projects. having stuff *local* to your computer, and already curated to be the information you're going to need, is really handy.

(deleted and re-made to fix the threading)

this is table stakes, and it's so, so frustrating that it's not the case on Linux.
fundamentally, Linux hardware compatibility is what it is because the commercial incentive of device manufacturers is to go after the largest audience and ignore the rest.
we're hopeful that the open hardware approach, as exemplified by the Pine64 people, can get more traction in the years ahead.
yep. it's a trade-off... we've been lifelong CLI users personally, and it's helped us a lot, but it was also a lot to learn the first time.
more thoughts on making scripting easier. neat idea.
writing to other people's screens on a shared Unix server was so fun.
it's nice to hear about existing projects that have already found the sweet spot
on hardware compatibility. yeah - this is an undeniable truth and we shouldn't look away from it. the question is what we do about it.

(yet another threading fix. sorry about that!)

it's true. there is no room for this kind of ego-driven conflict in the FOSS community we want to see.
every technology has its good and bad points. other people's tech doesn't take anything away from YOUR favorite tech just by existing. it's not about glory.
very much agreed. sorry, but current Unix shells could do with a lot of refinement. just because it's the command line doesn't mean UX is irrelevant.
this is a great observation! in the day, we remember the Linux Documentation Project put out some great guides that helped us and our friends a lot. no idea what's going on with them lately.
interesting. yeah - hackers in the movies always seem to have a tailored dashboard for whatever they're doing, which is so seldom the case in real life...
another vote for customization. for the record, we strongly agree.
hmmm, yeah. this is a very abstract problem, but an important one.
another tool that aims to make it easier to learn a config syntax
yeah... it's important to build those on-ramps to help people bootstrap the conceptual knowledge.
cosigned. Krita is a very solid image editor, it really shows that FOSS doesn't have to mean bad UX.
ooh, fun idea. kinda a new take on thin clients.
very much agreed! even KDE, which tries to present a unified settings UI, doesn't go as far as System 7 did in the 90s.
start with a block of marble, and remove all the parts that don't look like an elephant?
in the spirit of highlighting communities that already do this well, here's an example of a game that's highly hackable
it does have to be friendly to be fun, at least for most people
oh this is amazing! we had no idea things like this still existed.
we're going to link more of Nire's thoughts about how to reduce context switching and shorten feedback loops, because it's well-thought-out and important
having the vocabulary to think about a problem is a big deal. really affects how easy it is to find help.
definitely. we don't need all these quirky syntaxes. they aren't adding anything.
here's a great side thread about what it means to break concepts down into small pieces
we had not seen this! really cool. assembly language is a particularly fascinating area to try to make accessible, because it's so far beneath most people's awareness these days.
lots of people have weighed in to suggest tutorials, but a tutorial *about how to customize things* is a great idea.
the infrastructure for the "turn any computer into your computer" thing already exists, it's just a matter of making it accessible.
the whole "escape game" thing made us realize that it would be great to teach kids how to subvert or disable the surveillance software that parents and schools love to install. disobedience is an important part of life, sometimes.
as a queer person, we look at this surveillance stuff and we think: how many kids won't even survive to adulthood, because it took their safe haven and turned it into a nightmare?
there is a very good reason that so many queer people are awesome hackers. we HAD to be.
yeah, agreed. there's a complicated balance here, because corporations will absolutely take over everything if the culture doesn't reject that, but we shouldn't tie that to elitist attitudes or punish people for using the stuff they want to use.
we should be sprucing up the command line a bit. just some new decorations, that sort of thing :)
yes! this ties into a point we've made before, about how things stop being FUN when they become WORK.
oh yeah! we played that, it was neat.
yeah, definitely. we've written before about how, for example, we think professional creative tools should be maintained by user-owned co-ops.
other types of software may need other funding models, it really depends a lot on who it's trying to help and what it does. communication platforms, we think, are too important to allow any form of centralized control - and that sucks money-wise, but it's the constraint we have.
ah! thanks for asking this, we're sure you're not the only person wondering. "UX" means "user experience". basically, UX is the study of figuring out why users are having trouble doing things and making it better.

there is an actual, data-driven science to UX. it's its own specialty, separate from programming, in fact much closer to graphic design in terms of the core skills.
not all UX is about graphics; a lot of it is about understanding what mental models users have, and then changing the software to support those metaphors. that's an idea that even applies on the command line.
free software tends to be bad at UX, because free software tends to be the hobby project of someone whose core skill is programming.

it is what it is; it would be nice for FOSS communities to recognize the broad set of skills that are needed to make good software.
we're personally confident that command-lines *can* be given good UX, but it certainly will take a lot of work. also, of course, everyone has a different idea of what "fun" means.
yes, we personally agree that the 80s microcomputer ecosystem was amazing, and that this was a lot of why.
<3 all our love to Mozillans, and we do personally believe that it's a meaningful protection to have the non-profit structure that they have there, but it isn't magic, it has its limits.

it's very true that corporations have made huge inroads into taking over what used to be a wild and free community, and chopping it into tiny pieces that serve the interests of profit above people.
co-signed. speaking as a plural system who are also autistic and bipolar, and have various physical disabilities as well, it is incredibly important to build tech communities that are inclusive.
to be clear, we're huge fans of the command line, we think it can be very empowering. it's more versatile than a GUI can ever be. we think there's room for GUIs too though. some people will prefer them, and that's okay.
the counterpoint is also important to keep in mind. both these perspectives are true. neither "wins", they should be held in dialectic with each other - always be thinking about how they both apply.
absolutely. Apple understood this before anyone, and we think more Linux developers would benefit from reading all the material they've put out about human interface considerations.
not that we think people should pigeonhole themselves as "Linux developers" vs "Apple developers" or whatever. it's important to keep a broad perspective.
your intuition about what your users understand is probably wrong.
huh! we hadn't heard about Puppy Linux. interesting.
yeah - we agree with this, too. different UI paradigms have different affordances - they make different things easy or hard.

even so, some people will prefer to do the "worse" thing because it's a better fit for how they personally think, in all their human weirdness and glory.
very much agreed. this is a great goal to aspire to, and current systems fail it badly.
yeah, spreadsheets are a really interesting example.
how to get started with Puppy Linux, mentioned a few tweets up

(sorry! we re-made this tweet because we broke threading. again.)
yeah we definitely want more creative explorations like this. the past is still worthwhile, but we should always feel free to re-mix it in new ways to build the future.

we're going to step away for a bit and do other things, but please keep chatting about all this! it's been so cool seeing everyone's perspectives
we want to make sure that this bit from the middle of the thread, about our motivations for asking this question, is easy to find for people arriving late.
one final thought to close out this thread, because we know that not everybody has gone through the mental transformation of learning the activist mindset ("be the change you want to see in the world", and all that)...
... if you're a developer reading this, or someone with community-building-experience, and you see an idea in the thread that makes you think "that's really cool, someone should do that!"
YOU can do that!!!

we're always happy to chat if you need advice about the best way to try it.

• • •

Missing some Tweet in this thread? You can try to force a refresh
 

Keep Current with Irenes (many)

Irenes (many) Profile picture

Stay in touch and get notified when new unrolls are available from this author!

Read all threads

This Thread may be Removed Anytime!

PDF

Twitter may remove this content at anytime! Save it as PDF for later use!

Try unrolling a thread yourself!

how to unroll video
  1. Follow @ThreadReaderApp to mention us!

  2. From a Twitter thread mention us with a keyword "unroll"
@threadreaderapp unroll

Practice here first or read more on our help page!

More from @ireneista

5 Dec
there's a lot of people who are taking firm stances on this latest anthology that's been going around trans circles, and it's just, like...
we realize you're convinced of your stance. also, many of you were just as convinced of the *diametrically opposed* stance last year when Isabel's story was published.
try not to be mired in the moment like this. try to think about whether the objections that seem so real right now, ... whether you'll even remember them next year.
Read 7 tweets
4 Dec
The clouds in front of the mountains are pretty today. They only come halfway up, which is always an interesting look.
Vancouver's city planners went to great pains to preserve lines of sight for this stuff, and it paid off.
The water in the harbor makes different textures in different places, and it's always neat thinking about what causes that.
Read 5 tweets
31 May
The immediate implication of the Amazon mesh networking thing is that some people won't know where their bandwidth is going. This is concerning in its own right...
... but to us, the larger implication is that once it's deployed, it will no longer be possible to prevent a device from reaching the internet just by not connecting it to the internet.
To make this concrete, we'll give an example.

If you don't want your Roomba to publicize a map of your home, today, you can be sure it won't by never setting up the app or the wifi. (Not that you should have to go to that length!) nytimes.com/2017/07/25/tec…
Read 8 tweets
31 May
if reading the words of the Voight Kampff test summons up a big chunk of emotions, pride swirling with fear...

well, there's a good chance you're autistic or otherwise neurodivergent, and had to deal with some crap from the medical system as a kid.
it's kind of emotionally incomprehensible to us that there are people who have lived their entire lives and never had somebody seriously interrogate, with high stakes, whether they are a person or not
it's unclear to us whether Blade Runner was intentionally commenting on psychiatry, but it's a very clear comparison
Read 4 tweets
24 May
So there's a thing we want to describe about toxic masculinity and how it flourishes. We think it's important for people on the left to think about because the counters to it are only obvious when you understand what they're for.
The pattern we're going to describe is not a universal one, there are certainly exceptions to it. Also, it does come up occasionally in other contexts beyond toxic masculinity. We're just trying to set the stage by describing the place we see it most.
The pattern is this: People who are part of this culture tell each other, early and often, that they are all worthless.
Read 38 tweets
23 May
when you're trying to build a sense of group identity from nothing, for a demographic who until very recently have been isolated, never knowing each other... there are a lot of challenges #pluralgang
there is no single "plural culture", nor will there ever be. ask one system what their culture is, and you'll get five opinions.
the culture we get is the culture we all build, together.
Read 18 tweets

Did Thread Reader help you today?

Support us! We are indie developers!


This site is made by just two indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member ($3/month or $30/year) and get exclusive features!

Become Premium

Too expensive? Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal

Or Donate anonymously using crypto!

Ethereum

0xfe58350B80634f60Fa6Dc149a72b4DFbc17D341E copy

Bitcoin

3ATGMxNzCUFzxpMCHL5sWSt4DVtS8UqXpi copy

Thank you for your support!

Follow Us on Twitter!

:(