there's probably some extra bandwidth usage but it's using Data Channels, not streaming video. i'd love to know what a single character footprints is in ssh & in webrtc data-channels. evidently ssh is not actually that great[1]! webrtc might perhaps do better!! it is sctp/data-gram based! (personally i hope webtransport/quic eventually replaces this kinda odd/big stack)
once one starts doing Terminal User Interface programs or something is streaming to the screen, i expect potential differences to get less noticeable. hoping that packets get bigger. but definitely would like to see real evidence that packets get bigger, buffered, not just more numerous.
If you like this, you might find my little Tmux plugin[0] useful - it's meant to make continuing your terminal work on-the-go as simple as possible:
- you're in tmux
- you do some keyboard shortcut
- you're presented with a QR code
- you scan the code with your phone
- your active tmux pane is now in your phone, you can see output and send commands
It's targeted at working on remote hosts. WebRTC is baked in and most of the processing is done in the frontend app. For example, when you scroll, instead of sending the key to the server and waiting for tmux to update the display it all happens in the frontend as fast as possible.
This is the second time in a couple of days I've seen a character terminal implemented on top of the modern web platform. I am unable to insert a graphic of a withered void wojak in this post, so you will have to imagine it.
what was the other? personally i've embraced our new web tech overlords in many ways normally native, including finding chrosh on chromeos fit my needs just fine. although lossy UDP/WebRTC for terminal rendering sounds... a bit terrifying.
I love being able to bike around, whip out a fold up bluetooth keyboard & small phone mount, & hack away whenever I feel like it. Most rides I don't do anything. But being able to, whenever inspiration strikes, start coding is amazing. My phone has superb DPI, and it's well positioned at eye level, closer to my face than a laptop screen; I hardly lose any capability with this ultra-easy-packing on-the-go setup.
The biggest problem is that I haven't figured out how to remap caps-lock to control/escape on Android, but I'm hoping someday to have a real Linux/FreeDesktop phone.
Also see yesterday's article from replit[1], on different code editors. It talks a lot about who uses code editors & how much happier their mobile users are with some editors over others. Do you learn dev topics on your mobile? Wouldn't you like those learnings to be more interactive, not just static blog posts & articles? Things you can experiment with? Sounds great to me. Ok, so we'll need mobile capable editors. Or perhaps it's just that your family just can't afford a PC for you. It feels like your experience of not having tried implies that you think we shouldn't care, that this is irrelevant; I'd challenge us to keep an open mind. To think of the possibilities & uses broadly. To be welcoming & excited.
I don't really study dev topics on mobile. I do see this as being a pretty cool idea if I was outdoors camping or biking and needed to make a hot fix for my company. Personally if inspiration hits me when i'm out, i would make a note of my thoughts in notion and move on. I rarely sit down to code for less than an hour block of time.
ok fine. if you want to apriori rule out any potential uses for yourself that's fine. you do you.
personally i deeply value interactive mediums, even while mobile. fine that you dont. great that you have you your own one & done application that's all you need. oh to be completely well served & in want of nothing.
generally i feel like you've been more interested in self selecting & rejecting in all but narrow self defined circumstances. you permitted one ultra rare ultra circumstantial situation. you do you. but this is how is reads to me. it reads like a big rejection.
Meh not sure what the big deal is with "self selecting" and "self defined circumstances". Perhaps you should make more arguments for it so I don't have to but you do you.
That circumstance I presented isn't actually as rare as you might think. It was a genuine use case for myself and actually opened my mind from thinking it was completely useless, to thinking it is a niche.
> Meh not sure what the big deal is with "self selecting" and "self defined circumstances". Perhaps you should make more arguments for it so I don't have to but you do you.
You're not sure, but self selecting/self-defining matters a lot to me. Whether we merely apply our own pre-dispositions & already-knowns as to what is, what we know, & regard that as what shall be, or whether we admit that we don't fully know the possibilities that are about & haven't really tried outside possibilities matters a lot. To me. You can be meh about it if you want, but knowing our unknowns is vitally important to me. One needs to try to have experiences, in my book, before they are adequately rejectable. Even if it doesn't work for me, understanding how it works for another is just as important. Good experience sharing relies on empathy & openness, in my view; that's how we begin to figure out what foundations we may or may not personally share.
> That circumstance I presented isn't actually as rare as you might think. It was a genuine use case for myself and actually opened my mind from thinking it was completely useless, to thinking it is a niche.
I'm not sure what scale of current, existing plurality we need to start ignoring & telling those other people that their points of view & engagements are relevant. You're basically saying you are the status quo here, and I don't really disagree. I think that's probably accurate. Probably isn't rare. Where we seem to differ is that you think that matters a lot. I don't think that really matters for trying to read & understand the value latent here.
We don't really know yet, we can't really assess the genuine value of this. There's definitely some segment that stands to be helped immediately, strongly. But if this tech really is as valuable as it feels like it could be, it will be a changer, it will ongoingly beget new & different experiences. Me and you are both limited in our crystal ball reading on this. Neither of us really knows what potentials are afoot here. Even if we don't end up seeing radical vast & different change, we still won't really know what potential there was here. There's all manners of good reasons to believe this isn't significant, that it won't change things. But the absence of evidence is not evidence of absence of value. For this basic Pascal's Wager belief, I think it's worth spending more effort trying to explore why we might opt to believe in this tech, to wonder about it's possibilities, instead of reflecting on why we think it doesn't readily apply to current circumstances. Not that we should reject counter-though. But it seems pre-emptory, since we have so much left to explore about why-for, which will inform strongly what why-against concerns are most important to look at.
My original reply to you was actually me agreeing with you and slightly modifying your use case for me. How is that self selecting or self defining?
> I'm not sure what scale of current, existing plurality we need to start ignoring & telling those other people that their points of view & engagements are relevant.
22 comments
[ 0.15 ms ] story [ 75.9 ms ] threadonce one starts doing Terminal User Interface programs or something is streaming to the screen, i expect potential differences to get less noticeable. hoping that packets get bigger. but definitely would like to see real evidence that packets get bigger, buffered, not just more numerous.
[1] http://blog.hyfather.com/blog/2013/04/18/ssh-uses-four-tcp-s...
[0] https://github.com/bjesus/muxile
The biggest problem is that I haven't figured out how to remap caps-lock to control/escape on Android, but I'm hoping someday to have a real Linux/FreeDesktop phone.
Also see yesterday's article from replit[1], on different code editors. It talks a lot about who uses code editors & how much happier their mobile users are with some editors over others. Do you learn dev topics on your mobile? Wouldn't you like those learnings to be more interactive, not just static blog posts & articles? Things you can experiment with? Sounds great to me. Ok, so we'll need mobile capable editors. Or perhaps it's just that your family just can't afford a PC for you. It feels like your experience of not having tried implies that you think we shouldn't care, that this is irrelevant; I'd challenge us to keep an open mind. To think of the possibilities & uses broadly. To be welcoming & excited.
[1] https://blog.replit.com/code-editors https://news.ycombinator.com/item?id=29569711 (1 day, 0 comments)
personally i deeply value interactive mediums, even while mobile. fine that you dont. great that you have you your own one & done application that's all you need. oh to be completely well served & in want of nothing.
That circumstance I presented isn't actually as rare as you might think. It was a genuine use case for myself and actually opened my mind from thinking it was completely useless, to thinking it is a niche.
You're not sure, but self selecting/self-defining matters a lot to me. Whether we merely apply our own pre-dispositions & already-knowns as to what is, what we know, & regard that as what shall be, or whether we admit that we don't fully know the possibilities that are about & haven't really tried outside possibilities matters a lot. To me. You can be meh about it if you want, but knowing our unknowns is vitally important to me. One needs to try to have experiences, in my book, before they are adequately rejectable. Even if it doesn't work for me, understanding how it works for another is just as important. Good experience sharing relies on empathy & openness, in my view; that's how we begin to figure out what foundations we may or may not personally share.
> That circumstance I presented isn't actually as rare as you might think. It was a genuine use case for myself and actually opened my mind from thinking it was completely useless, to thinking it is a niche.
I'm not sure what scale of current, existing plurality we need to start ignoring & telling those other people that their points of view & engagements are relevant. You're basically saying you are the status quo here, and I don't really disagree. I think that's probably accurate. Probably isn't rare. Where we seem to differ is that you think that matters a lot. I don't think that really matters for trying to read & understand the value latent here.
We don't really know yet, we can't really assess the genuine value of this. There's definitely some segment that stands to be helped immediately, strongly. But if this tech really is as valuable as it feels like it could be, it will be a changer, it will ongoingly beget new & different experiences. Me and you are both limited in our crystal ball reading on this. Neither of us really knows what potentials are afoot here. Even if we don't end up seeing radical vast & different change, we still won't really know what potential there was here. There's all manners of good reasons to believe this isn't significant, that it won't change things. But the absence of evidence is not evidence of absence of value. For this basic Pascal's Wager belief, I think it's worth spending more effort trying to explore why we might opt to believe in this tech, to wonder about it's possibilities, instead of reflecting on why we think it doesn't readily apply to current circumstances. Not that we should reject counter-though. But it seems pre-emptory, since we have so much left to explore about why-for, which will inform strongly what why-against concerns are most important to look at.
> I'm not sure what scale of current, existing plurality we need to start ignoring & telling those other people that their points of view & engagements are relevant.
I am just confused about what you mean here..
edit: On closer inspection of the video at 1080p it actually uses parrot.live[1], so it's an OK tool ;)
[0] https://github.com/jmhobbs/terminal-parrot [1] https://github.com/hugomd/parrot.live