Hacker News new | past | comments | ask | show | jobs | submit login
Life at 800 MHz (artemis.sh)
260 points by blackhole 6 days ago | hide | past | favorite | 256 comments





> Technically Gentoo is also in the running, but can you imagine trying to compile all your packages from scratch on a system that benchmarks worse than a raspberry pi 3?

Uh, I actually did this, it wasn't so bad honestly it just took about a day to rebuild everything.

Honestly the Sony VAIO that I had was _awesome_ in some regards, the hi resolution display was extremely crisp! It fit comfortably in my inside jacket pocket, the battery didn't suck.

The only issue I had honestly was the proprietary connector to get ethernet (though this was more annoying in 2012 when I was doing this, these days laptops don't seem to have ethernet); the only other issue was that the GPU was extremely slow with Linux.

it was probably extremely slow in Windows too, but vista (which was installed on the thing) was far-far too heavy to understand why it was slow at all.

The nearest best laptop I've found that is in all areas superior than the Sony VAIO P-Series (aside from being a bit taller) is the GPD P2 Max which is basically perfect.... if only it had a passively cooled ARM CPU.

https://gpd.hk/gpdp2max2022


Compiling your own software is a really humbling experience. When it takes way more time to compile a browser than a full fledged OS or you find out that seemingly simple programs need to pull a mind boggling amount of dependencies you really start to question the state of the software world

I think the main reason browsers are so extremely slow to compile is the heavy templating.

But, I agree, I can compile my entire OS including user-space software and desktop environments in about the same time it takes to compile chrome.

Which is scary.

But then again, people want it to do everything (WebUSB, WebGL etc; etc; etc;). So it stands to reason that it's inherently complicated and difficult to compile.

I wonder if the high iteration time hampers development...


> I wonder if the high iteration time hampers development...

You might be interested in this post from someone on the Edge team at Microsoft:

https://textslashplain.com/2020/02/02/my-new-chromium-build-...

In particular:

> I returned to Microsoft as a Program Manager on the Edge team in mid-2018, unaware that replatforming atop Chromium was even a possibility until the day before I started. Just before I began, a lead sent me a 27 page PDF file containing the Edge-on-Chromium proposal. “What do you think?” he asked. I had a lot of thoughts (most of the form “OMG, yes!“) but one thing I told everyone who would listen is that we would never be able to keep up without having a cloud-compilation system akin to Goma.


> ...that seemingly simple programs need to pull a mind boggling amount of dependencies you really start to question the state of the software world

Recent jawdrop: 'apt-get install asciidoc' on a pi needs to pull 189 packages, will use 889Mb of additional disk space.


Maybe you already know, but in case not or someone else needs this: try with --no-install-recommends, it skips a lot of bs.

I don't recall exactly what it was, but I remember installing something like a tiny library and it wanted to also install mysql-server or something like that >_<


is good advice for any system not operated by a novice

  echo "APT { Install-Recommends \"false\"; }" >/etc/apt/apt.conf.d/90norecommendsthxby

Why is that not good advice for a novice too? I do this by reflex every time I install Debian or Ubuntu, and to my experience it did not create a situation that needed "expertise in apt".

due to depends on latex and friends.

Gentoo was fun, too bad I don’t have time for it anymore. I used to go for nice walks when Firefox was compiling. Great opportunity to go outside and take a break.

USE flags in Gentoo also allows for a much more configurable system.


I use a GPD Micro PC throttled to 6 watts TDP, which means the fan can stay off permanently. It fits in a jeans back pocket, and has an ethernet port. And a serial port. And a full size HDMI port. And three full size USB ports, and a USB-C port.

I wouldn't trade it for much...


Well you'd trade it for around $600. That's what it costs, new.

AKSHUALLY

https://en.wikipedia.org/wiki/Endowment_effect

(but also, what I meant was there are few extant devices I would prefer to be using, even if price were no object. It's nearly perfect.)


Details please, which? Or is there a whole range of options?


>>> imagine trying to compile all your packages from scratch on a system

used to be the norm back in the unix days. finding exact pre-compiled binaries for your exact arch/OS combo was like finding a pot 'o gold ;)

am also amazed at how well gba emulators run on older devices!


You just reminded me of this site I used to use for Sparc/Solaris binaries:

https://www.ibiblio.org/pub/packages/solaris/sparc/

I’m sure it’s been 15 years since I’ve even thought about it.


I also did this (around ~2008), a friend of mine and I built near identical Atom boxes with first gen (Diamondville) 64bit atoms on Intel motherboards running 865 chipsets IIRC. The GPU/Chipset was louder than the CPU because the CPU was completely passive. I did emerge Xorg on that... it took I think a day and a half(ish) even optimizing the heck out of compile options to use everything march=native... it was slow as heck. But it lasted me for years as a little project box until I replaced it with an 4th gen i5.

You really do start to ask yourself if you need a package if compiling it will take a day or two. Hence OpenOffice never got installed.


> GPD P2 Max

Save for the processor being better than any VAIO's, I disagree. I find all of these to be absurdly unreliable (crappy firmware) and very cheap hardware for the price, not comparable at all to the typing experience on the P-Series. And the "trackpoint substitute" is a disaster, resembling a "tiny touchpad" more than a trackpoint.


Hrm, interesting.. I disagree with your opinion about the hardware quality, it feels sturdy and keys travel quite nicely. The screen is fantastic in color reproduction (for my needs), has high resolution and gets bright enough.

There’s no trackpoint/nipple and I hadn’t considered that a problem as I’m weird and spent a lot of time getting used to only using the keyboard some years ago- so an oversight on my end and you’re completely right, the touchpad sucks.

The firmware is extremely bare bones, but I wouldn’t say it sucks since I don’t have any reason to believe it’s bad. (Nor good, it just works for me.)


I think you are mistaking the GPD Pocket 2 Max with the non Max version. The Max version has a real touchpad while the GPD Pocket 2 has that tiny thing on the upper right.

I have the GPD Pocket 2 and while the trackpoint thing is not the most precise you get used to it and I have no complaint to do regarding the hardware. It just works.


At one point in college, I was using an old Thinkpad x41 tablet and wanted to mess around with gnuradio. I wanted it on my tablet laptop since I had that on me most hours of the day. Compiling gnuradio took several hours. I was running arch so I want unfamiliar with compile times for things I grabbed from the AUR but it was atrocious. I started it in my first class of the day and would just throw my laptop in my bag while it was still compiling and walk quickly to my next class so I could grab a power outlet before Pentium M sucked up all the battery.

I did Gentoo on a 600mhz Athlon. It was certainly a humbling and informative experience.

Glad you enjoy your life at 800 MHz! I appreciated your article although the plural form to address a single person (not the editorial "we") makes me uneasy for political considerations.

So many more things could be easily enjoyable on such hardware if the software ecosystem allowed it. I'm also curious what hardware modularity like Framework is doing could have achieved two decades ago: if you could easily plug in a chip to decode/encode video quickly, this computer could probably play any kind of video.

> We have no idea what crates.io thinks it makes sense to require javascript to look up packages but here we are.

I've had a similar experience with crates.io:

    curl https://crates.io/
    {"errors":[{"detail":"Not Found"}]}
Apparently, without a specific Accept header, crates.io thinks i want a JSON response for a crate lookup, not the homepage. Now i don't even remember why i was requesting this URL to start with (not in a script) but i don't understand the logic of that and the maintainers in the chatrooms seemed to consider it's not a bug.

I'm also very curious about antiX "proudly anti-fascist" distro but that they're two debian releases late (still on stretch) does not exactly attract me.


I don't want to speculate about Artemis specifically, but first-person plural pronouns to refer to oneself typically isn't a "royal we" or anything like that, it's just what helps some folks feel comfortable, especially those who have DiD or who label themselves as plural. See https://www.reddit.com/r/plural/wiki/index (keywords: "plurality," "multiplicity," ...)

I'm dating someone who refers to themself in the first person plural; it becomes perfectly natural pretty quick :)


I never knew this was a thing. I'm not on board with promoting the use of "we" as a replacement for first person singular as being an acceptable societal norm, unless you're the Queen.

Sorry, but it is too close to contributing to mental health, or personality, disorders for me.


> I never knew this was a thing. I'm not on board with promoting the use of "we" as a replacement for first person singular as being an acceptable societal norm, unless you're the Queen. Sorry, but it is too close to contributing to mental health, or personality, disorders for me.

Wait until you find about about languages like Hindi where the plural form can be used for respect even when referring to an individual :)

I hadn't realized that calling an individual in plural was even a point of contention until comments on this thread pointed it out (likely because I'm used to it from Hindi). Don't forget, the author may be bi/multilingual.


> Wait until you find about about languages like Hindi where the plural form can be used for respect even when referring to an individual :)

In fact it’s pretty common amongst a lot of languages. Most Latin-derived languages use the plural to show respect. But of course, never to talk about yourself. You’ll use the pluralized form when talking to strangers or to people who are over you hierarchically (but this usage tends to disappear in a lot companies).

As a French, reading someone speaking about itself as "we" is shocking not because it looks like there is multiple people involved (but it also does) but because it looks like the person tries to be "above" you hierarchically. Of course i know it isn’t what’s intended but language interpretation is an automatic mechanism.


> As a French, reading someone speaking about itself as "we" is shocking not because it looks like there is multiple people involved (but it also does) but because it looks like the person tries to be "above" you hierarchically. Of course i know it isn’t what’s intended but language interpretation is an automatic mechanism.

That's interesting, culturally. In India, it's very common for example for people from North India to be much more "pride-based" where individual identity is important and people often use the plural for themselves, while in more southern states there's lesser emphasis and singular is much more common. A lot is dependent on culture.


> Wait until you find about about languages like Hindi where the plural form can be used for respect even when referring to an individual :)

I speak German, which uses uses the plural sie/polite Sie.

It's completely different to "we" being used by an individual to refer to themselves in English.

The fact that a language construct might exist in other languages is irrelevant. Calling a girl "it" in English would similarly be bizarre, although that is the grammar in German.


Something cannot be a disorder unless it causes harm. Things that are not disorders and are out of the ordinary can be considered adaptations and can be advantageous.

Harm to whom? Many things can cause harm to oneself (socially, at least) without harming anyone else. Being odd about your pronouns is one of those things.

If you ain’t changing and if you ain’t adapting you might as well not be alive :p

I may as well not be alive if I'm not adapting to people using "we" as a personal pronoun?

What is there to "adapt"? Whether or not I use "we" to refer to myself doesn't affect you, and also doesn't ask you to change your behavior.

> Whether or not I use "we" to refer to myself doesn't affect you

It affects anyone who reads the article, as it's incorrect english and makes it harder to interpret.


Pedantically, there is no such thing as "correct" or "incorrect" English as there is no standards body that dictates such rules. You can verb any word you want and people will understand what you mean.

You don't have to read the article if it doesn't meet your muster. Why bother caring?


Who's to say it's "incorrect English?" You clearly understood what was said. This person's identity seems like the least interesting and most irrelevant part of this discussion.

Who cares? This is such a trivial thing to pay attention to.

> I never knew this was a thing. I'm not on board with promoting the use of "we" as a replacement for first person singular

Thou art fighting a losing battle; the grammatical first person singular will soon be as passé as the second.


After finishing the article, my main take-away was how impressive it is that such a quirky tech setup could work for both of them. I was comparing it to my relationship and how difficult it is to share any item/space which is also customized to either of our preferences. It gave me hope.

Then I read these comments.


I grew up with the ~10 MHz 8086 PC, and I was on bulletin boards and the Internet around the 486 era, still stuck in the "tens of megahertz" era. Even wireframe 3D rendering at 640x480 was glacially slow. CAD applications on a CPU without a floating point unit were just unbelievably painful.

800 MHz and a solid state disk is luxurious if you're not wasteful with it. As the article's author points out, this is "not up to you" ("we"/"us") any more, other people get to decide how much JavaScript to shovel on top of web applications.


But what is the intended purpose?

It seems ambiguous to me, I was honestly trying to figure out if there was more than one person using the author's laptop, or if it was a multi-author article or something.

Not that English isn't chocked full of ambiguity - I just haven't managed to identify a benefit over using the more commonly accepted "I" here.


I interpreted this as the "editorial we" or perhaps the "author's we":

> The editorial we is a similar phenomenon, in which an editorial columnist in a newspaper or a similar commentator in another medium refers to themselves as we when giving their opinion. Here, the writer casts themselves in the role of spokesperson: either for the media institution who employs them, or on behalf of the party or body of citizens who agree with the commentary. The reference is not explicit, but is generally consistent with first-person plural.

https://en.wikipedia.org/wiki/We

It's quite standard usage.


> label themselves as plural

Definitely contributed to me not finishing the article.


Life honestly becomes so much more pleasant when you avoid interacting with the pronouners as much as possible.

I find that sad. Do you also hold that view when interacting with people different in other ways? People that dress differently, hold different political or religious views, people from other places to you?

I have acquaintances who was lyrical on the topic of LGBTQ+ folks, feeling that somehow, non-LGBTQ+ folk are being "oppressed". I keep asking them, "What are they taking from you? What could you do before that you're not allowed to do now?". They typically don't respond or change the topic. We both know that the types of behaviours that are no longer "acceptable" lie on a spectrum that starts with "being casually disrespectful", to outright *ism. It's sad to me that a lot of people value the rights of some to be jerks, over the rights of others to partake equally in society and to feel equally safe and valued in public spaces.


Oh I’m very much supportive of living your life as you see fit. But it’s the pronouners who insist on you using they/them (grammatically silly) or who insist on sharing their pronouns when it’s obvious what they are, then attempting to guilt you into sharing yours even though they know what yours are. This behavior tends to bleed into other annoying personality traits as well.

And yes I do tend to avoid other annoying types of people too, across the spectrum.


> or who insist on sharing their pronouns when it’s obvious what they are, then attempting to guilt you into sharing yours even though they know what yours are.

Don’t worry, you’re already using singular “they” without even realizing!



This argument is always brought up - "it was used centuries ago, so it's still valid now!"

Except: all normal human beings trip up on it; those who believe in the they/them BS trip up on it; and almost no one who uses this argument supports other things that were done centuries ago, so it's not really arguing in good faith.

Arranged marriages at very young ages were a thing centuries ago, should we bring those back?

If you want to use custom pronouns, nobody is stopping you. The rest of the world is just annoyed and tired of hearing you desperately announce it every time we see you and we're not going to play along.


I wasn’t using it as an argument for reverting to some past usage, just pointing out that language and grammar evolve. Just like it evolved away from that usage, it may evolve back.

I find the fact that some people are massively triggered by this fascinating. Personally I’m happy to use whatever pronouns people desire for themselves if they make it clear to me. I get it wrong occasionally cos I have a lifetime of doing something different, but if someone has explicitly made their desire known to me, I’ll make the tiniest effort of referring to them as they wish. It’s not a chore for me. I think of it as being polite to that person.

If I tell you my name is Mike, and prefer to be called that, would you insist on calling me Michael or Micky or Mickster? Even if I told you I was uncomfortable with that (for my own reasons which I don’t have to share with you)?

> every time we see you and we’re not going to play along

Im guessing “we” is not referring to yourself here.


Ugh. Hacker News has traditionally been one of the worst places to discuss queer politics, but reading this has made me feel so frustrated that I can't help but weigh in.

> I find that sad. Do you also hold that view when interacting with people different in other ways? People that dress differently, hold different political or religious views, people from other places to you?

Yep! Ready to get nihilistic? Their existence is pretty much inconsequential to me. Sexuality, gender identity and appearance has quite literally zero bearing on the way I address other people. Unless someone make a concerted effort to be my acquaintance, I will likely forget about their existence within the hour. That doesn't mean I can't sympathize; but the internet has greatly distorted our idea of how important other people actually are. We conflate identity with politics and alliances, we grok importance by follower count and Google search results, it's a disgusting mess that can only be effectively deterred by not caring.

Is it sad? Hard to say, but I certainly feel like it's a less frustrating way to live your life when compared to bending over backwards for everyone. I operate with my own interests at heart; as much as I despise Ayn Rand's philosophy, she wasn't wrong when she said that the greatest minority is the individual.

> I have acquaintances who was lyrical on the topic of LGBTQ+ folks, feeling that somehow, non-LGBTQ+ folk are being "oppressed".

I don't think it's hard to sympathize with that sentiment, even though I'm a gay man myself. I feel embarrassed by the level of entitlement that the rest of the community seems to push, in public and online. A decade ago, the LGBT movement was pretty cut and dried - queer people wanted to integrate into society as normal individuals, without any pretense or opportunity for judgement. In response, they became a protected class and everything was pretty much solved. There hasn't been a legitimate reason to be mad as a gay person since those bakers refused to make a gay wedding cake, and even that only incensed me because it was against the law. As far as I see it, the modern LGBT movement is far too infatuated with liberties that don't exist, and hunting boogeymen that don't care. It makes me ashamed to be queer and wish that I could live in a world where my only identity didn't boil down to "the gay guy".


I have heard it from a "sovereign citizen". They seem to use it when wanting to talk about themselves (flesh) inclusive of their various personhoods and corporate entities. I imagine that traffic cops find it unsettling for a lone driver to say "we" are going somewhere, as if there are other people somewhere unseen in the vehicle.

> label themselves as plural

No


I have a laptop from 2009 or 2010 running at 800 mhz with a 32 bit CPU. It has to run an older version of Ubuntu (18.04) because nothing supports it nowadays. Even 32 bit packages are hard to get. I see no reason to use antiX or other esoteric distros since ubuntu runs fine on it and supports the hardware. I doubt antiX supports more hardware.

Someone else recommended it here, but I don't see the advantages over a robust package repository like ubuntu 18 or a minimal ram only distro like puppylinux. https://cheapskatesguide.org/articles/antix.html

Funny enough I got puppylinux running from a dos (windows) partition and running out of RAM on just 2gb on a Toshiba Portage m200. I've even got Windows XP Tablet edition running on SSD, but it can't really connect to much online due to the TLS limitations. And newer versions of the linux kernel don't support the wireless chipset. It is also difficult putting an old non-PAE kernel into a newer distro.

TLS really killed the utility of a lot of older computers with regards to using the "modern internet".


I have an old Dell with a 32-bit 2.33 Ghz T2700. Linux fully supports the GPU, and no issues with missing 32-bit packages on OpenSUSE Tumbleweed. It's a spare browsing / retro gaming machine hooked up to the TV in the guest room. For gaming, it runs everything from arcade MAME to Mario Kart 64 like a champ. For browsing, it's not speedy but not bad on heavy HTML sites like gmail/youtube.

I agree antiX was a poor choice. No issue with PAE kernel on Tumbleweed i686. If OpenSUSE ever drops x86 support, there's always Debian or Arch 32 (if I want to stick with a rolling distro).


I still have a tablet PC from 2005 in rotation, and the lack of 32-bit apps is definitely a killer, but not terrible.

My original reason for reviving it was for use as a whiteboard in Zoom calls, but there's no 32-bit Zoom app - and I'm sure screen sharing while decoding 15 people's video would've been out of the question anyway. So I run a VNC server on it, and share out a VNC session from my work laptop instead.

I've also hit the issue where I've had to compile software for x86 using modern build toolchains. It takes forever, and more often than not, I run out of RAM (only 1GB). To get myself out of a pinch, I've mounted a 16GB USB 2.0 flash drive as swap space. Sure, it makes compiling even the most basic software a multi-hour process, but where this machine isn't my daily driver, it's still easier (to me, at least) than cross-compiling.


Openbsd will run fine, even with TLS.

Sure it runs, but will it run TLS 1.3? that seems to be a big requirement for websites these day

Yes, no issues.

Do you happen to know if the Chromium port is packaged for this platform? Have you used it?

It's packaged, and I tried it. But you should boost up /etc/login.conf values on resources.

> I'm also very curious about antiX "proudly anti-fascist" distro

"Anti-fascist" doesn't actually mean that - it's a political dog-whistle.

> they're two debian releases late

That's in line with their use of Palemoon, which lags behind normal Firefox feature (and security) releases due to their decision to support older features (mostly XUL) (not that this is very avoidable, because maintaining an XUL fork is very hard work, and not for the faint of heart).


Dog whistle for what?

There's no unified "anti-fascist" movement, but the common theme among the self-described anti-fascists I know is the belief that physical violence has a legitimate place in democratic processes.

Frankly they remind me of a line by Nietzsche about staring too long into an abyss.


> the belief that physical violence has a legitimate place in democratic processes.

That's not exactly the point, though. We are not in a democratic process (unless by democracy you mean giving away powers to congress), and our society is very violent towards the most vulnerable segments of it.

Do you think giving back just a tiny portion of that daily violence we face is immoral or wrong? How is it justified for people to threaten us with guns if we don't pay rent to some arbitrary landlord or to detain us if we dare steal food for basic survival, yet attacking bank windows or punching an actual genocidal nazi in the face is seen as violent?!


I guess fascists who want to prove that they can render their swastika on an Antix machine no matter what the developers do.

Communists and anarchists who are willing to use force against their opponents. (this is based on both media reports and first-hand experiences in Portland, Oregon)

If you mean against the ruling class threatening a million species and the neonazis promoting eradication of many branches of our species, then yes i'm certainly advocating to stop these people by any means necessary.

> > We have no idea what crates.io thinks it makes sense to require javascript to look up packages but here we are.

>I've had a similar experience with crates.io:

They do have an API (ps: I built crates.live on top of it). I think they have a very good reasons to block the crawling of their main website. Otherwise, people might abuse it. Actually, they recommend you identify yourself when crawling their API to not limit you. I didn't do it, and found no problem constantly calling their APIs.


> Otherwise, people might abuse it.

So first, "crawling" a website is not abusing it. It's simply using the website and there's nothing wrong with that. Then, i believe that "not found" JSON message was not intended as an anti-scraping measure, but was in their view a meaningful error in the sense that i did not request info about a specific crate so the API responded "not found".

What's weird is no specific Accept header, http://crates.io returns 301 with some HTML, and https://crates.io/ returns 404 with some JSON, while in a browser you get a proper 200 with HTML. I just found that pattern very confusing, but hey maybe i'm just an old dinosaur and that is the future of web development.



AntiX is a wonderful systemd-free debian and now also has Sid. It's fast as fuck in usage. It's best left as a live system.

Sounds like that guy "All" from Zoolander

> Glad you enjoy your life at 800 MHz! I appreciated your article although the plural form to address a single person (not the editorial "we") makes me uneasy for political considerations.

You assume Artemis identifies as a single person. In all likelihood, they are a plural system. Statements like yours are microaggressive at best.


A plural system? You say that as though it requires no explanation.

I don't get the "political considerations" part, but this is the first time I've encountered anyone referring to themselves as "we" online, and I also found it jarring.


A plural system is multiple identities or personalities in one body/mind. Plural systems are increasingly demanding to be recognized and respected as such -- and companies are starting to comply. Much like trans and nonbinary identity, plurality is an aspect of identity we're all going to have to deal with now.

Genuine question: How do you differentiate this from full-blown mental illness? Because this sounds 100% like what society traditionally recognizes as schizophrenia/split personality disorder. Or, in more extreme cases & phrased less politely, insanity.

schizophrenia != split personality! I think what you mean is https://en.wikipedia.org/wiki/Dissociative_identity_disorder.

Does the United States current recognize plural folks as a protected class? Do we even have the infrastructure to recognize them in any meaningful fashion? To extrapolate on that, how much research has gone into understanding the dysphoria that these people experience? Do we have a medical basis of understanding when it comes to how plural systems affect the mind? Do we even know if it's healthy to address plural systems as their individual components?

I apologize in advance if this sound antagonistic, but putting plural identities on the same levels as queer and trans ones seems... a little premature, if you ask me.


you don’t just decide to be plural, it’s caused by mental illness. acting like multiple people is not being plural.

There was one slip in the article though: "did I say". Which makes it even more jarring to read, IMO.

Ouch, right in the start.

They started using that pronoun only 3-4 months ago, so it still slip in. I'm more puzzled why 'we' but not 'our'?


maybe I'm just an ignorant Eastern European, but this honestly boggles my mind. every medical handbook would classify this as mental illness, no?

A girlfriend of mine had surgery on both wrists.

She got the Dragon Speech software, and I was surprised at how good it was.

You can of course dictate all your notes, documents emails. It also provides means to navigate your OS, start programs, close them, and a lot more.

It is expensive but she could do most of her work with two hands that didnt work.

A while back I saw a video about a guy who wrote code using such software (not sure what he used in particular). This can be tedious "Open bracket", "new line" etc.

He had spent a long time tuning it so it was fast and efficient. He used a set of custom grunts and noises as "macros" for all the bracket brace, and other symbols that are in heavy use in programming languages.

If you were just listening to him and didn't know what he was doing it sounded a bit distressing.

https://www.nuance.com/dragon/businesbs-solutions/dragon-pro...


You refer to Tavis Rudd's PyCon 2013 demo: https://youtu.be/8SkdfdXWYaI

I write code with speech to text, and it's nothing like this.

Anything that's can be templated is. There's natural language integration with LSP. I use Vim mode "naturally" etc...

It's not like reading what's on your screen word by word. It's less input than typing.


To add to that:

You'd use a custom vocabulary as well. So rather than "curly open" you'd use "heck", and instead of "enter" it would be "bark". I'm just making the actual words up here, but the point is to use a different/more simplified vocabulary that's also easier to understand by the computer.

https://talonvoice.com/ is also worth keeping an eye on.


I'd love to hear a short sample of what this sounds like!

Simple example of mine:

  def tree_size(n: Node) -> int:
    if not n:
      return 0
    return 1 + tree_size(n.left) + tree_size(n.right)
would be:

funk tree size takes near type cap node returns int slap

if op not near next return zero slap zing

return op one plus call tree size pass near dot left

op plus call tree size pass near dot right

commands like "funk", "op" and "if" insert snippets

slap = end enter, zing = shift tab, next = move to next snippet placeholder, near = letter n

still a work in progress as I find ways to make it flow better.


I wonder with copilot would you just say "new function called" and it would make your block statement.

what software did you use?

I worked with a guy who wrote code like this. He was, indeed, pretty productive, but it was hell sitting next to him without good headphones. Was this guy you're referring to a long haired, kinda scruffy guy who had worked at Amazon at one point?

Did your GF or friend ever consider using foot pedals at all? I knew a programmer once who used various foot pedal combinations for different punctuation marks and tabs.

yes.

I have been looking into pedals before this ever started, and we looked at some different options, but could not find something that seemed worth it.

I really want a set of foot controls to act as my mouse since growing a third arm is currently not practical. I keep looking around and I know there are some solutions out there, but not in my price range that seems solid.


Have you considered buying a USB gamepad and putting it on the floor?

404 link

https://www.nuance.com/dragon/business-solutions/dragon-prof...

I wonder if the parent comment typed that URL by hand.


I suspect a laptop touchpad, because the touchpad might have buttons at the top, and the extra "b" could be close to them.

A cozy laptop sounds nice. I bet IRC is more than fast enough, surprised it didn't get a mention. Also, if you just want to read some text on the web as fast as possible, w3m might be worth a shot. I use it in TTY2 all the time to look stuff up. Browser CDN caches like Decentraleyes or LocalCDN might also be worth trying especially with the mnestic set up: you would only have to load certain JS bundles once per session.

>a dishonorable mention to twitter for being slower than Discord, we wish we were making that up

If you're just browsing Twitter, then the Nitter frontend (https://github.com/xnaas/nitter-instances) is way, way faster. Does not have algo-recs either, which could be positive. If you need to post, I assume you've tried spoofing user agent to mobile? This might help with bloated sites in general.


Check Bitlbee, you'll have IRC proxies for everything. Twitter, Slack, Telegram, anythiing Pidgin supports with the -purple build.

For music, mocp, and links+/dillo make a good combo.

Youtube-dl+ytfzf+mpv with a config setting up the youtube-dl format for 420p = heaven.

In ~/.config/mpv/config:

         ytdl-format=bestvideo[height<=?420]+bestaudio/best
For the rest, Fluxbox+rox+lxappeanrance+nm-applet+xpdf. Ted and Gnumeric as a micro office-suite. Or Siag, if you don't need Unicode.

On Chromium, it has a --light switch.


The 1000x480 resolution seems interesting. Maybe this machine would make a good single-purpose device for writing.

Also, somewhat related: Former Debian maintainer Joey Hess famously used a Dell Mini 9 for all his coding [1, 2]. I wonder if the Sony has a better, less cramped keyboard compared to the Mini 9.

Another interesting guy doing valuable work on low-end, underclocked hardware is Nils M. Holm [3].

Myself, I can get most of my stuff done on a Thinkpad T42 (underclocked to 600 Mhz to reincarnate its dying GPU). With the ram-booted Tiny Core Linux, this thing still flies. I'm having a hard time ditching it because of the 4:3 IPS screen and excellent keyboard. I've even used it to produce lengthy radio programs for my country's public broadcasting.

Aside web browsing, there seems to be more than enough software solutions, hacks, workarounds and programming languages for doing valuable work on rather old hardware these days. Really interesting times we're living in.

Then again, might be true that with yesterday's hardware, you're limited to solving yesterday's problems. I guess I'm fine with yesterday's problems in many aspects of life.

1: https://news.ycombinator.com/item?id=4721645

2: https://joeyh.name/blog/entry/xmonad_layouts_for_netbooks/

3: https://usesthis.com/interviews/nils.m.holm/


Some more great musings on actually using low-level hardware (inspired by Nils M. Holm's work and setup): https://news.ycombinator.com/item?id=18292613

Some people doesn't know that, aside of media creation and consumption, we don't need so much power to do other things.

Most of my university assignments were done on a Acer Aspire One netbook (1.3/1.6 GHz Dual Core Atom, 2 GB DDR2 RAM) and I had no problem. To program in C, C++, and Python in Debian is simple great, and to simulate circuits with SPICE related software on Windows 7 is also good.

I started using it because it was more light and more comfortable than the newer laptop I had (15" 4th gen Intel i5 laptop), and as a small device for reading PDF is great, so i ended up using it more and more, and for more tasks, leaving it for exclusive academic usage and letting the other for games and media.


My Acer Aspire 1 is still kicking with an external monitor.

It was supposed to be a disposable laptop, it outlasted and persisted through everything else.


I still have my Samsung NC10, had it running an IRC bot until recently in power save mode with no fans. Opening a modern version of a browser is pretty revealing about how heavier the web has become though.

The NC10 was/is a great machine. Considering the dimensions, it had a remarkably good keyboard. I also liked the "fanless mode". It felt quite sturdy, and, iirc, you could open the screen all the way down, to 180 degrees. The one I had for some time did suffer from its symptomatic "white screen" issue, though.

I did some writing on this machine, and I always felt really concentrated, quite possibly because of the small screen.


Seconding this. The NC10 was an amazing little thing for writing a lot when on the go. Too bad the white screen issues have killed most of them off by now.

I have the first EeePC, still working and with a replaced battery. It makes for an adorable little ssh terminal.

Haiku OS works well on them, too.

As for media creation, SaaS is where its at for weak endpoints. My ancient chromebook battery is going and it could never run CAD, office, or video editing natively, but it runs onshape which is SaaS 3-d CAD, and Google Workspace/suite/apps whatever its called this week, and Wevideo SaaS video editing perfectly fast no slowdowns or problems pretty much ever. The onshape viewer works great on my phone and tablet so if I'm building something far away from my desk, I've got the prints with me. Unlike my desktop keyboard, my tablet touch screen is sawdust-proof.

Another discovery I made a long time ago was network connections are usually fast enough and small battery friendly CPUs are slow enough that its faster to send a video file to AWS (or have it there to begin with), spawn a linux box on AWS, run handbrake in CLI mode to convert the video to some obscure format on a very CPU beefy machine, and download the converted file, and delete the huge (and expensive) AWS instance, than it is to transcode video locally. Some CPU based transcoding is very slow if you don't have a lot of cores and its brutal thermally and to the battery.

If you only have one SaaS app in your life, the old meme was what do you do when the internet is down? Well, the internet is almost never down for me, I'd pick up my laptop and go to a cafe or library if it was, and everything I do is online or SaaS or VPN'd in so I wouldn't crabby about one app being down I'd be crabby about being completely and totally shut down.

That anti-SaaS argument in 2020's is like arguing that people have to drink bottled water because what would they do if tap water stopped working one day? If we're in a situation where the tap water stops working then we got bigger problems than which bottled water company to enrichen.

The linked article seemed surprised that a 2009 device could play video, but I had been using Mythtv for 7 years by that point including occasional HD video on a relatively weak settop box class of computer and doing youtube for awhile so his specs for playback seem very low compared to what I was doing in '09 on small devices, but whatever.


Doesn't OnShape actually run client-side?

What is the allure and purpose of going back to 800 Mhz? I mean I did it myself this week, but was frustrated enough to think it's a really dumb idea, waste of time. I can't even articulate why I did it in the first place.

I used a Raspberry Pi 4 (1500 Mhz) as a daily driver for 4 days. Struggled with hidpi scaling, no Signal Messenger, overheating CPU, Youtube at 360p, HTML Gmail.

I went so far to upgrade Pi to SSD, plus heat sink. Considering adding active cooling... but the said nope, back to Macbook Pro. Why do we even try?


Change your workflow. You cannot expect a less powerful system to perform the same as a more powerful system.

Rather than watching YouTube directly, use youtube-dl with VLC. Rather than using HTML Gmail, use IMAP and a native email client. Rather than using Eclipse, use vim.

We all fall into patterns. We grow to find comfort in those. But, we can't expect to maintain those patterns when circumstances change.


Thanks for the good suggestions. I'll try them out if I ever find a reason to try again!

> Why do we even try?

Depends on how hard you want to try or compromise on.


Sorry, I have to ask about the pronouns. Does the use of "we" imply that this laptop is shared by multiple people?

I also found it confusing. I was wondering if it was this person's preferred pronoun but their Twitter [0] lists "she" as of "January 2022" and all the testimonials use "she" too [1].

[0]: https://twitter.com/EverfreeArtemis

[1}: https://artemis.sh/


The article is written to be read back in Gollum's voice

Nah, the author is royalty and is referring to the use of the laptop by themselves and by their sovereign station.

It's possible the author has multiple personalities, it's a good thing that they are running a multi-user operating system.

I mean, I suppose it's a reasonable pronoun for someone to use if they wish to be referred to as "they" ...

"They" is almost exclusively singular in those circumstances, regardless of its etymology. Similar to how "you" derives from the old English second person plural pronoun, but is in virtually all variants of modern English acceptable for the second person singular.

By definition it does. I'm not sure what weird literary device the author is going for here.

No, it's just a pronoun flex. Spend as many cycles on that as you think it deserves.

"So this thing’s main job is to help us stay off our phone, since touch screens are the hardest on the health of our hands."

I have never heard this before. On the other hand I have heard about keyboards being an issue many times. Anybody else know anything about touchscreens being harder on hands than keyboards?


I can only speak for myself but I find using a smartphone upright one-handed for longer than a few minutes uncomfortable because of the bizarre positions it forces my fingers and palms into. It's a flat, slate-like object and our hands are designed to grip round things that protrude into our palms. I also need to keep my thumb free to use the touchscreen and my hands are on the smaller side so I end up balancing it on a protruding pinky and contorting the inside back edge of my hand. It's no problem for me to use two hands or hold it in a screen-up orientation but holding it in front of me for more than a minute isn't fun (despite having decent dexterity and grip strength from deadlifting, typing, playing instruments, etc).

I do wish landscape typing hadn't gradually gone by the wayside. I used to be a firm landscape-typer and found it much more comfortable for my hands, but I've accepted that phones just aren't designed for it anymore - too often the keyboard fills too much of the screen to see the textbox adequately.

I sometimes tend to let my phone fall into the nook of where my pinky finger meets my hand, when I want to hold it near-vertically/upright. I can agree that this doesn't really work well long-term, it's much more comfortable to hold it flat since it's less likely to simply fall out of my hand (...) that way.

I occasionally brace my index finger against the top edge of the display; this used to work great on my Note 3 with its giant bezel (particularly at the top), my current Mate 20 Pro's notched edge-to-edge screen doesn't play well with this though :(


I defintely find "large" phones to be very uncomfortable and cramping to use for more than a few minutes. Thats why I haven't upgraded from my first gen iPhone SE. Thats even a little too big--the iPhone 4 and previous iPhones were the perfect size. I'm not a heavy smart phone user, but for the two weeks I had an XS, it was the most difficult phone I ever had. I couldn't hold onto it, and was dropping it constantly. So I gave it to my brother and bought another SE used (before the XS, I had an SE that got water damaged.)

If you are holding a mobile phone in a hand your palm and fingers (thumb) are more or less in the same position. Take a look at https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5546699/, https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7440311/, https://www.toi-health.com/physician-articles/effects-smartp... and https://www.sciencedirect.com/science/article/pii/S000368701... just to name some articles.

None of the studies you linked concluded any significant findings. The most significant, which was unremarkable, was from the first:

"There is limited evidence that MTSD use, and various aspects of its use (i.e. amount of usage, features, tasks and positions), are associated with musculoskeletal symptoms and exposures. This is due to mainly low quality experimental and case-control laboratory studies, with few cross-sectional and no longitudinal studies."


That tripped me up for a few minutes. I decided that since the author is using we in the singular not plural so it is that person's experience only and not meant to be a blanket statement. Also, I can only assume that the author meant a touch screen on a phone and not, say, a touch screen on a laptop because I can't imagine how that is difficult on someone's hands.

"our" is not the royal we here, this person refers to themselves with a plural sounding pronoun for identity reasons. They are not making a statement about ergonomics generally even though of course I can see how that can be confusing.

There's nothing really hard on healthy people when we use touchscreens, but this person clarifies they are disabled so this sounds like an edge case for them because if you're disabled in some ways, a nice fat keyboard is just going to be a lot more gentle on your hands and fingers than the tiny thumb keyboards of mobile devices.


Additional context: this is a person who wrote a brainfuck interpreter in sed on my couch using an iPhone. I tend to trust their input as to typing comfort implicitly.

That's hilarious (and very impressive), but also not applicable to the vast majority of humans that use touch screens! I mean, just how much do people type on touch screens?

That touchscreens are not conducive to significant typing seems like an additional point against them (and this matches my experience).

They're too powerful for glass.

Yeah, I hit that sentence and was hoping for more details.

Smartphones are hell for thumbs and index fingers

Out of curiosity I checked eBay to see the going rate for this particular portable; of the two listed, one has a starting price of $350 and I watched the other go from a $99 bid to $150+ in an hour. Apart from the quirkiness or the need to replace one's recently dead machine, I can't wrap my head around such a high price for such low performance. For a little more than the higher priced unit, one can get a Gemini PDA or similar device with a more modern and faster processor, and come out even more portable and with excellent battery life (though I did note the author's need for a non-touchscreen device due to a handicap, the touchscreen on a modern portable doesn't have to be used if there's another pointing device).

I have one. Out of curiosity, mainly.

I paid just under £100 for mine, about 3Y ago, and another £20 for the VGA/Ethernet dongle -- which doesn't seem to work, FWIW.

It's a fun toy but not much use. Lovely form factor, poor keyboard, poor trackpoint (and I like the things), and sluggish.

I wish they did a modern one with a better keyboard, though.


This is more along the lines of my “vintage”.

Thanks kind of feel left out when folk here start remembering their c64 and Ataris and whatnot!

My first computer was a celeron 500MHz with windows 98 (maybe there was a 300MHz with win 3.1 but I never got it working)

So, this blog is nostalgia! Winamp and the Linux clone!? DDR2!? Back in my day we had some other thing that I don’t remember the name (sdram?), we ruled the city because with winrar we could use the T1 of the university to download stuff, then split it in 4 3.5” 1.44Mb floppy disks to install on our computers!

Oh, and CD-R changed the game forever! And usb… it took a while and a few dongles (parallel to usd, serial to usb, ps2 to usb) and hunting down the proper .inf file, but it was glorious!

That’s my kind of nostalgia :)


>So, this blog is nostalgia! Winamp and the Linux clone!?

XMMS: http://www.xmms.org/


That one!

I got my first computer in the days of CD-ROM and was amazed that a CD could hold more data than my Win95 (later Red Hat 6 (not RHEL)) Pentium Packard Bell's 512MB HDD could.

"Thanks kind of feel left out when folk here start remembering their c64 and Ataris and whatnot! My first computer was a celeron 500MHz..."

In a lot of ways, between the c64 and the celeron 500MHz is an enormous, almost unrecognizable leap, whereas between the celeron 500MHz and the machine in my hand is just a lot of incremental change. I had a machine ~2000 that was de facto a 500MHz Duron (running at its full 1GHz overheated very quickly), and I used the same basic paradigm on that as I'm using now. Emacs, browser, terminal windows, MP3 player. Wifi needed a dongle. The integrated webcam is new since then.


My first computer connected to internet was the family Pentium 100Mhz with 64MB of ram that ended up retired by my father for a much powerful athlon whatever. It used to run Windows 98. I inherited that one and immediately ditched the Windows 98 as it couldn't allow me to listen to music while browsing and use an office apps, audio was stuttering all the time.

This is actually the reason I started using linux. I remember internet was usable at 100Mhz back in the days, I could play some videos (obviously at a much lower res, I was using a CRT). The funny things is some of the apps mentionned here already existed at the time so it resonnate with my experience, although back in the days I tended to prefer apps running on terminals unless it was absolutely necessary. Emails, music player, I was a terminal for all of that. My computing life was not that different than today bar the videos resolution increase. And the web wasn't less interesting or usable.

It is incredible how crap internet has become that we can't reasonnably think we can browse it comfortably on what would have looked like a supercomputer at the time of my 100Mhz Pentium.


>I used the same basic paradigm on that as I'm using now. Emacs, browser, terminal windows, MP3 player.

I don't disagree with your overall point, but two parts of that paradigm, Emacs and the terminal emulator, do date back to the C64 era. Here's Richard Stallman on developing GNU Emacs: "There were people in those days, in 1985, who had one-megabyte machines without virtual memory. They wanted to be able to use GNU Emacs. This meant I had to keep the program as small as possible." [0] Emacs may never have run on Commodore computers, but my impression is that it ran on very similar computers.

[0]: https://www.gnu.org/gnu/rms-lisp.html


MOS Technology 6510 @ 1.023Mhz here!

I cannot fathom…

My closes reference is a PIC16F that I used to program in 8bit assembly, but the thing was 8MHz, and I only blinked an LED! :)


Closer than you think since IIRC the PIC, like the Z80, takes 4+ clock cycles to complete one instruction and the 6502 can sometimes do it in 1 (albeit a much simpler/limited core, but obviously Commodore/Apple/Nintendo et al made it work).

They made it work by not doing picture rendering on the CPU. Same as GPUs today. Still impressive though

https://www.youtube.com/watch?v=XTRkZ-SKs5g

That's basically the CPU running at 1.024 Mhz. The video hardware is dumb, runs independent of the CPU, and just scans a region of memory to send pixels to the display. All software pushing pixels otherwise.

You are not wrong with the NES, C64 and other machines using a graphics chip with sprites and other hardware features to assist in various ways. But, quite a lot happened on the CPU.

BTW, this game is done on a 1Mhz 6809, all software pushing pixels.

https://www.youtube.com/watch?v=wAKxa5C9jHY

(I would skip out to the middle somewhere in that video to get a sense of what is being done)

On that game specifically, it's a single frame buffer. Screen divided into two halves, each drawn while the display is delivering the other to the player.


The Fujitsu FM-7 line of 8-bit computers actually shipped with two 6809 compatible CPUs (Hitachi 6309 IIRC) and the second one just did software graphics the whole time pretending really hard to be a GPU.

Dual 6309 chips? NICE!!

I won't bother, because I've got a good retro hobby going and under control, but I want one of those. :D


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: