A much better alternative to this would be to for more terminal emulators to implement overstrike mode (DECSET 20, as available on the VK100). That way you could just draw border characters over existing text cells to achieve the same thing.
Unfortunately mintty is the only terminal emulator I'm aware of which supports overstrike [0], as this would open up a world of new possibilities for TUI apps if more widely supported.
I’m not against overstrike, but I can think of one drawback of that solution: it becomes significantly harder to have rounded corners with a filled interior.
But I’m not trying to argue for nor against either solution.
There's some downsides to this - I can't recall where I called this out off the top of my head, but this also adds a bunch more things that that wouldn't do that are covered in the spec.
If you mean your proposal adds more things, that sounds like it might be over-complicated. To solve the problem of tightly-fitting boxes around terminal text -- if you consider that to be a real problem -- overstrike sounds like a really elegant solution.
Which things do you consider in your experience of writing tuis that are mentioned as being wanted in the blog post / rationale doc linked to be unnecessary?
If you want those things then overstrike is insufficient.
cell borders sound even more elegant imo, they are a targeted, precise fix for a useful feature. overstrike sounds messy to me though I can see where it has a lot of uses.
Compatibility will always be the hard part with new standards like this. Given that most terminals already have some special handling of box drawing characters, I don't know if it's completely dead on arrival for implementors, but I don't love as an application author with ratatui, not knowing how the padding on my app will look. I might honestly rather disable this and know it'll be more consistent, than enable it and fragment my UI.
The ratatui change was a pure PoC, isolated branch, my own fork, not the ratatui repo even. This would likely never ship as a default approach as it would be way too breaking. I haven't really spent any time on making it hook into the current block border approaches - it's just at the cell level where this is opted in. That will never merge without much much more though into things.
I used ratatui there mostly so that I could prove that it would work in real looking apps that do panels and similar stuff that benefits from borders.
But yeah, compat is key. It would probably go out to ghostty, rio, kitty pretty easily, terminal.app never, other terminals much slower. IDK about the windows / linux side of things.
I love this work and everything else Ratatui does but I have to ask: if you're changing the terminal protocol and the terminal itself to do this, why not do something more ambitious; a full monospaced information-dense UI framework, a terminal designed for user interfaces, rather than one designed to replicate the limitations of dumb terminals?
I've written about it before, but in brief the mountain of standards is worth ripping out wholesale and doing write as a semantic protocol. It's needed to solve several things that work really poorly in terminals because their unit of compatibility is cells rather than a higher primitive. It's hard to bake something on top that works well.
The need is most strong in Accessibility, Text wrapping, Scrollback handling (i.e. making the scrollback editable and not dead text), and various other bits that make writing highly functional TUIs. Coding agents tend to find all those edge cases.
This often gets push back that it's easy, to which I can point nay sayers at the various many months long issues on things in Codex, Gemini, Claude etc. on flickering, scrolling, key handline, IME, ... doing anything moderately complex involves modeling a terminal in a terminal app effectively.
Ghostty and Rex sort of do the opposite bit. If you think of it like a stack, they write a new layer underneath that transports the cells on a better multiplexer layer 0. I think a new semantic layer 2 is an approach that would be neat to work with.
I think I mostly just want to see someone do it so I can see what it looks like; it just feels like something that should exist. You're taking baby steps towards it! :)
I kinda keep talking about it in the hopes that someone will be nerd sniped enough to throw tokens at it as an idea. Sort of what you're probably doing too I guess. ;)
... Write up a more complete spec of the problem and I could be tempted to throw some tokens at it. But before I do, can you write up a more complete description of the problem with existing terminal protocols? I don't think I understand the issues well enough to make a good solution. Eg, can you give some examples of what apps are hard to build now? What prior art is worth looking at? Etc.
> a full monospaced information-dense UI framework, a terminal designed for user interfaces, rather than one designed to replicate the limitations of dumb terminals?
I can only speak for myself but I use the CLI and TUIs specifically because of the limitations those modalities impose on software that uses them as the primary method for interfacing with the end user. If you want a full featured UI and all that comes with it why not just use QT or GTK or modern HTML or etc?
I think the only "improvements" I'd like to see are thing such as metadata regarding what the different streams are for, an official way to communicate using structured data as opposed to raw bytes, that sort of thing. Spiritually in the same ballpark as the difference between HTML before and after CSS.
I actually am exploring exactly this. Ironically from the perspective of someone that was trying to leave the Terminal.
How i got here i find kinda interesting. Originally i was trying to leave the terminal. Using GPUI/etc frameworks. I kept running into the terminal though, because no matter how nice the UI is - the terminal still needs to work excellently.
Eventually i decided to try an abomination. To use a custom terminal which, not unlike Kitty images, allows generic apps to pass escape codes to control the UI components. Where the UI components have no restriction on "dumb terminals". So my text editor is a two part app, an editor and a tty to render the actual custom UI APC codes.
This paved the way for true smooth scrolling from within a terminal editor, as the editor effectively paginates but the tty scrolls within the buffered pages. It allows for different UI components to mix fonts and font sizes, eg popup docs being a different size than the other text, and .. well, anything i suppose.
I have to say it's working excellently and is slowly becoming my daily driver. I won't link it though, it's a fully vibe coded and even claude designed horror.
I'm surprised i've not seen more of this in the past to be honest.
I am running ewm (not exwm) which makes emacs into a wayland compositor and it's fantastic, most of the primitives dissolve and you script it with elisp, wayland windows are just a special buffer and where applicable normal windowing and buffering rules can be applied.
The developer seems to have used AI heavily in the elisp code and it works very well.
> To use a custom terminal which, not unlike Kitty images, allows generic apps to pass escape codes to control the UI components.
> I'm surprised i've not seen more of this in the past to be honest.
This is what emacs is. It's a text buffer, but text can have properties like how to render, and custom values. then a bunch of commands to manipulate the buffer and it's contents.
You can save all that effort and instead come up with a CSS framework that enforces monospaced fonts, uniform font sizes, encourage information-dense layouts, have borders that don’t take up a cell (border: solid 0.1rem black; padding: 0.5rem;). Pair it up with a small JavaScript framework that listens to keypresses and buffers them. Furthermore, the framework makes the user decide on fonts, whether to allow images, whether to adhere to a strict character grid etc.
For a while in the 2000s I tried programming with Verdana in my IDE at the time. It wasn't all that bad. It would be in a terminal however - fun idea regardless.
Because it will turn into the kitchen-sink mess that the Web is.
Terminals work so well because there's a huge frictional cost to adding to the protocol. That's what keeps the protocol simple, which is what keeps multiple implementations feasible. Unlike the web, where unless you have GoogleBux you can't afford to implement it properly.
Text terminal with every keystroke round-tripped through the server is not a design that makes sense for any reason other than backwards compatibility. It's kind of annoying to design for and atrocious for performance.
What I'd really love to see instead of all of this is an OS that has composable areas that can interact and have event loops of their own. E.g. ditch the app boundary and compose. We're getting there sorta with AI Canvas type things (e.g. github copilot app, some of the stuff in chatgpt visualizations), but all of this sort of work tends to devolve to browser type things instead of simpler components. This is mostly unrelated to the terminal side of things though from my perspective on this. But it's a fun thought experiment.
This is pretty cool but this does feel like we are just reinventing another GUI display at this point.
I don't generally like TUI's and think one gets more flexibility with raw text interfaces and being able to glue things together is more important then them looking nice. I think less fancy out put is better to be honest.
I am in general against the big push to fit GUI interfaces into the tty, but I have to say that this feature looks genuinely good. Vim will look great with this.
I appreciate the innovation happening with TUIs in these modern times. But, aesthetically, I have to say, I like my borders to have some breathing room, which gets lost in this particular treatment.
The big thing to note on this is one last caveat. I don't care about the actual spec on this - happy for that to be whatever works. I care more about the capability than how it's done / implemented.
There's some bits that feel like they probably make sense to consider doing other ways, like how border placement works (inside, outside, on the border, etc.), and how the filling works - to make borders incorporate external or internal padding.
Why wouldn’t this work over tmux? The authors claims don’t make a whole lot of sense because if this were a cell property, like any other CSI, then it’ll get rendered in the terminal emulator exactly as intended. And we know this because none of the other CSI sequences get garbled over multiplextors.
But for there to be a dedicated section on how multiplexors wouldn’t work suggests to me that the authors already tried this via tmux.
So I’m left wondering what they’ve done with the implementation of this to make that a problem.
I spotted they are using DCS sequences. But even that shouldn’t be an issue itself.
Tmux drops DCS that it doesn't know like this. That aside, this is extra state per cell that needs to be tracked, handled, understood from a rendering perspective etc. It can work, but would require maybe a thousand lines of code or so to support storing and handling state, redraw etc. and all the various edge cases.
I upvote naysayers comments there as I don't think votes should be used to indicate (dis)agreement at least not without commenting why, more an indication that this is a useful thing to discuss. So if you see my comments not getting votes while every other one has a bump, that's my vote on them :)
It's kind of funny, I posted this here and there at similar times. Expected to be spam flagged there due to the anti-AI brigade, but my submission here was the one that actually got hit (likely because I disclosed AI use in the post body is my guess).
Thanks for that slop warning, on a general point: wondering whether there might be enough momentum to finally ditch the whole terminal garbage and use proper GUIs in spite of what the card-punching devils intended?
Sure - I use GUIs most of the time despite maintaining Ratatui. I much prefer them for coding agents as the amount of interactions and various things I want out of the interface are not handled by TUI well. GUIs are a vastly superior interface (I worked on the Codex TUI at OpenAI btw).
Nothing really - Ghostty has a bit of padding that's unused per cell - 16 bits. The implementation stores an ID into a lookup table with border combinations. If all of those were used and never cleaned up, 65535 at around 100 for each combination used, order of magnitude 25MB absolute worst case assuming no real optimization in place for this.
Having markdown tables render with zero-cell spacing (use underlines for horizontal and different background shades for vertical) is just one of the billion reasons I will never, ever, ever go back to closed-source coding agents.
I think the other side of things is why have the graphic program run over ssh, just have raw text cli tools to manipulate what ever then have a program driving it. This for me is where emacs excels and some programs already fit well with this model and the fantastic (albeit hacky) tramp.
Doing it through port forwarding is kinda bad though because generally access to network ports is not very well access controlled. Doing it through unix sockets (like x) would be much better but annoyingly browsers do not support http over uds
Afraid that’s a contradiction. In this role, HTML is a pile of hacks upon hacks that was not originally intended for UIs.
But, yeah, people are trying to reinvent GUIs. By layering hacks over TTYs. If only there’d be some protocol for widgets and events over a stream connection…
> In this role, HTML is a pile of hacks upon hacks that was not originally intended for UIs.
I get where you're going and am genuinely curious, what does a "better" system designed to do what HTML/JS have become actually look like? (or what would one look like?)
There has to be markup of some sort, and there has to be scriptable interactions of some sort, right? If not, how else would it work?
the thing with X and wayland etc is, that they are low level. which is fine if you want full control over the presentation. what HTML was trying to do is to go higher level, send over some data and let the browser figure out how to present it. but it went wrong and people are abusing it as a low level presentation interface like X or wayland.
what we are missing is a really high level protocol/system that takes really high level data and figures out the presentation on its own.
take a list of files for example. you can run ls -l in a terminal and it will attempt a crude text based formatting of the output. or you can run a full gui filemanager.
what if there was a command that would produce ls output as a table data structure, run in a sort of terminal that knows how to format a table nicely without the ls command having to include any formatting instructions? this terminal would have support for all sorts of data types and structures to display all of them nicely.
> Take a list of files for example. you can run ls -l in a terminal and it will attempt a crude text based formatting of the output. or you can run a full gui filemanager.
ls's output (and a few other programs') are human facing. They're not meant to be machine readable.
> what if there was a command that would produce ls output as a table data structure, run in a sort of terminal that knows how to format a table nicely without the ls command having to include any formatting instructions? this terminal would have support for all sorts of data types and structures to display all of them nicely.
Because that's just complexity for its own sake. What's simpler is to just create a shell alias for a particular view you want. The shell is for interactive usage, with some capabilities for automation. If something better is needed, you write a real program for it.
> what if there was a command that would produce ls output as a table data structure, run in a sort of terminal that knows how to format a table nicely without the ls command having to include any formatting instructions? this terminal would have support for all sorts of data types and structures to display all of them nicely.
So, powershell? For some reason only IT jockeys on Microsoft's death grip remember and use its full capability after learning it.
UI in a browser is a massive hack in itself, a clean design would be better. For example a compact semantic remote access protocol to use with UIs, something like SPICE but better optimized for networks, fully presentation-independent, and less janky. This would let everyone use a GUI framework of their choice with an adapter. But people building terminal GUIs are culturally not ready for this, they'd rather continue reinventing the wheel and piling up hacks in terminals. I wonder when they realize that proportional fonts are more readable when vertical alignment is handled separately, and introduce terminals with variable-width fonts. And the presentation layer (I suggest calling it CSS).
A pet peeve of mine is that most people / WM get "useless gaps" wrong. You do want a gap between two terminals because otherwise there are times where things can get really confusing (where you don't know what belong to which terminal / window).
So the "borderless" (0 pixels) / zero "useless gaps" simply are confusing and... Wrong.
But then people who do use useless gap, as in TFA, get it wrong too: they typically add gaps near the monitor's edges.
Yeah. But no. Sorry but my monitor's edge is a perfectly fine gap. I don't risk mistaking it for anything.
So I modified my WM (the Awesome WM) to do proper "useless gaps" computation and remove the ones that really are useless: and that'd be those near the monitor's physical border.
I mean: nearly everybody gets it wrong, but it's not complicated to get it right, is it?
P.S: even the terminology is wrong... People calling those "useless gaps" haven't got a clue: they are useful. The only useless ones are the one on the monitor's edge (if your monitor has physical edges, which it probably does).
Man, I’d really like a monitor with no edges. One that just keeps going and going, past my window and the tree outside until it’s completely out of sight. 100% useful area.
One benefit worth calling out: the box-drawing fallback is worse than it looks for CJK users. U+2500-257F and the half/full block elements are East Asian "Ambiguous" in UAX #11, so in a terminal configured to render ambiguous as double width (xterm's cjkWidth, iTerm2's checkbox) every line glyph takes two cells and the whole frame lands in the wrong place. Cell borders aren't text, so that failure class just disappears.
The part I'd want specced early: borders are per-cell, but cells aren't per-grapheme. A CJK char owns two cells plus a continuation placeholder, and combining marks or ZWJ emoji sequences own zero-or-more depending on whose wcwidth table you ask, so app and terminal routinely disagree across Unicode versions. Today that shows up as a vertical bar one column off; with cell borders it becomes a line that stops short of the corner. Does the border attach to the cell or to the grapheme cluster?
A much better alternative to this would be to for more terminal emulators to implement overstrike mode (DECSET 20, as available on the VK100). That way you could just draw border characters over existing text cells to achieve the same thing.
Unfortunately mintty is the only terminal emulator I'm aware of which supports overstrike [0], as this would open up a world of new possibilities for TUI apps if more widely supported.
[0] https://github.com/mintty/mintty/wiki/CtrlSeqs#overstrike
Overstrike is underrated!
I've really enjoyed using it on paper: https://hughpyle.com/ASR33/
I’m not against overstrike, but I can think of one drawback of that solution: it becomes significantly harder to have rounded corners with a filled interior.
But I’m not trying to argue for nor against either solution.
There's some downsides to this - I can't recall where I called this out off the top of my head, but this also adds a bunch more things that that wouldn't do that are covered in the spec.
If you mean your proposal adds more things, that sounds like it might be over-complicated. To solve the problem of tightly-fitting boxes around terminal text -- if you consider that to be a real problem -- overstrike sounds like a really elegant solution.
Which things do you consider in your experience of writing tuis that are mentioned as being wanted in the blog post / rationale doc linked to be unnecessary?
If you want those things then overstrike is insufficient.
cell borders sound even more elegant imo, they are a targeted, precise fix for a useful feature. overstrike sounds messy to me though I can see where it has a lot of uses.
Expecting Terminal Emulators to support overstrike is a high bar even many text editors don't even support strikethrough. Would love to see it though
Compatibility will always be the hard part with new standards like this. Given that most terminals already have some special handling of box drawing characters, I don't know if it's completely dead on arrival for implementors, but I don't love as an application author with ratatui, not knowing how the padding on my app will look. I might honestly rather disable this and know it'll be more consistent, than enable it and fragment my UI.
The ratatui change was a pure PoC, isolated branch, my own fork, not the ratatui repo even. This would likely never ship as a default approach as it would be way too breaking. I haven't really spent any time on making it hook into the current block border approaches - it's just at the cell level where this is opted in. That will never merge without much much more though into things.
I used ratatui there mostly so that I could prove that it would work in real looking apps that do panels and similar stuff that benefits from borders.
But yeah, compat is key. It would probably go out to ghostty, rio, kitty pretty easily, terminal.app never, other terminals much slower. IDK about the windows / linux side of things.
I love this work and everything else Ratatui does but I have to ask: if you're changing the terminal protocol and the terminal itself to do this, why not do something more ambitious; a full monospaced information-dense UI framework, a terminal designed for user interfaces, rather than one designed to replicate the limitations of dumb terminals?
Yep,and I want to do that too :)
I've written about it before, but in brief the mountain of standards is worth ripping out wholesale and doing write as a semantic protocol. It's needed to solve several things that work really poorly in terminals because their unit of compatibility is cells rather than a higher primitive. It's hard to bake something on top that works well.
The need is most strong in Accessibility, Text wrapping, Scrollback handling (i.e. making the scrollback editable and not dead text), and various other bits that make writing highly functional TUIs. Coding agents tend to find all those edge cases.
This often gets push back that it's easy, to which I can point nay sayers at the various many months long issues on things in Codex, Gemini, Claude etc. on flickering, scrolling, key handline, IME, ... doing anything moderately complex involves modeling a terminal in a terminal app effectively.
Ghostty and Rex sort of do the opposite bit. If you think of it like a stack, they write a new layer underneath that transports the cells on a better multiplexer layer 0. I think a new semantic layer 2 is an approach that would be neat to work with.
I think I mostly just want to see someone do it so I can see what it looks like; it just feels like something that should exist. You're taking baby steps towards it! :)
I kinda keep talking about it in the hopes that someone will be nerd sniped enough to throw tokens at it as an idea. Sort of what you're probably doing too I guess. ;)
... Write up a more complete spec of the problem and I could be tempted to throw some tokens at it. But before I do, can you write up a more complete description of the problem with existing terminal protocols? I don't think I understand the issues well enough to make a good solution. Eg, can you give some examples of what apps are hard to build now? What prior art is worth looking at? Etc.
> a full monospaced information-dense UI framework, a terminal designed for user interfaces, rather than one designed to replicate the limitations of dumb terminals?
I can only speak for myself but I use the CLI and TUIs specifically because of the limitations those modalities impose on software that uses them as the primary method for interfacing with the end user. If you want a full featured UI and all that comes with it why not just use QT or GTK or modern HTML or etc?
I think the only "improvements" I'd like to see are thing such as metadata regarding what the different streams are for, an official way to communicate using structured data as opposed to raw bytes, that sort of thing. Spiritually in the same ballpark as the difference between HTML before and after CSS.
I actually am exploring exactly this. Ironically from the perspective of someone that was trying to leave the Terminal.
How i got here i find kinda interesting. Originally i was trying to leave the terminal. Using GPUI/etc frameworks. I kept running into the terminal though, because no matter how nice the UI is - the terminal still needs to work excellently.
Eventually i decided to try an abomination. To use a custom terminal which, not unlike Kitty images, allows generic apps to pass escape codes to control the UI components. Where the UI components have no restriction on "dumb terminals". So my text editor is a two part app, an editor and a tty to render the actual custom UI APC codes.
This paved the way for true smooth scrolling from within a terminal editor, as the editor effectively paginates but the tty scrolls within the buffered pages. It allows for different UI components to mix fonts and font sizes, eg popup docs being a different size than the other text, and .. well, anything i suppose.
I have to say it's working excellently and is slowly becoming my daily driver. I won't link it though, it's a fully vibe coded and even claude designed horror.
I'm surprised i've not seen more of this in the past to be honest.
I am running ewm (not exwm) which makes emacs into a wayland compositor and it's fantastic, most of the primitives dissolve and you script it with elisp, wayland windows are just a special buffer and where applicable normal windowing and buffering rules can be applied.
The developer seems to have used AI heavily in the elisp code and it works very well.
> To use a custom terminal which, not unlike Kitty images, allows generic apps to pass escape codes to control the UI components.
> I'm surprised i've not seen more of this in the past to be honest.
This is what emacs is. It's a text buffer, but text can have properties like how to render, and custom values. then a bunch of commands to manipulate the buffer and it's contents.
You can save all that effort and instead come up with a CSS framework that enforces monospaced fonts, uniform font sizes, encourage information-dense layouts, have borders that don’t take up a cell (border: solid 0.1rem black; padding: 0.5rem;). Pair it up with a small JavaScript framework that listens to keypresses and buffers them. Furthermore, the framework makes the user decide on fonts, whether to allow images, whether to adhere to a strict character grid etc.
Or even more ambitious - ditch the monospace
For a while in the 2000s I tried programming with Verdana in my IDE at the time. It wasn't all that bad. It would be in a terminal however - fun idea regardless.
everyday we get closer to building emacs
Because it will turn into the kitchen-sink mess that the Web is.
Terminals work so well because there's a huge frictional cost to adding to the protocol. That's what keeps the protocol simple, which is what keeps multiple implementations feasible. Unlike the web, where unless you have GoogleBux you can't afford to implement it properly.
Text terminal with every keystroke round-tripped through the server is not a design that makes sense for any reason other than backwards compatibility. It's kind of annoying to design for and atrocious for performance.
Teletypes and dumb terminals managed to implement local echo. You don't need to round trip everything.
As awesome as this is, but I can't shake the feeling terminals are becoming lightweight browsers...
What I'd really love to see instead of all of this is an OS that has composable areas that can interact and have event loops of their own. E.g. ditch the app boundary and compose. We're getting there sorta with AI Canvas type things (e.g. github copilot app, some of the stuff in chatgpt visualizations), but all of this sort of work tends to devolve to browser type things instead of simpler components. This is mostly unrelated to the terminal side of things though from my perspective on this. But it's a fun thought experiment.
I kinda like the examples without the native borders…
The charm of the terminal is that you can do whatever you want, as long as it fits in one cell.
This is pretty cool but this does feel like we are just reinventing another GUI display at this point.
I don't generally like TUI's and think one gets more flexibility with raw text interfaces and being able to glue things together is more important then them looking nice. I think less fancy out put is better to be honest.
I am in general against the big push to fit GUI interfaces into the tty, but I have to say that this feature looks genuinely good. Vim will look great with this.
I appreciate the innovation happening with TUIs in these modern times. But, aesthetically, I have to say, I like my borders to have some breathing room, which gets lost in this particular treatment.
The big thing to note on this is one last caveat. I don't care about the actual spec on this - happy for that to be whatever works. I care more about the capability than how it's done / implemented.
There's some bits that feel like they probably make sense to consider doing other ways, like how border placement works (inside, outside, on the border, etc.), and how the filling works - to make borders incorporate external or internal padding.
Why wouldn’t this work over tmux? The authors claims don’t make a whole lot of sense because if this were a cell property, like any other CSI, then it’ll get rendered in the terminal emulator exactly as intended. And we know this because none of the other CSI sequences get garbled over multiplextors.
But for there to be a dedicated section on how multiplexors wouldn’t work suggests to me that the authors already tried this via tmux.
So I’m left wondering what they’ve done with the implementation of this to make that a problem.
I spotted they are using DCS sequences. But even that shouldn’t be an issue itself.
Tmux drops DCS that it doesn't know like this. That aside, this is extra state per cell that needs to be tracked, handled, understood from a rendering perspective etc. It can work, but would require maybe a thousand lines of code or so to support storing and handling state, redraw etc. and all the various edge cases.
Would be cool to get borders and lines out of copy/paste
There's a nice thread about this on lobsters, but the author is getting piled on.
I upvote naysayers comments there as I don't think votes should be used to indicate (dis)agreement at least not without commenting why, more an indication that this is a useful thing to discuss. So if you see my comments not getting votes while every other one has a bump, that's my vote on them :)
It's kind of funny, I posted this here and there at similar times. Expected to be spam flagged there due to the anti-AI brigade, but my submission here was the one that actually got hit (likely because I disclosed AI use in the post body is my guess).
Love this. And the people who like more space can opt for that (this doesn't look things in).
Thanks for that slop warning, on a general point: wondering whether there might be enough momentum to finally ditch the whole terminal garbage and use proper GUIs in spite of what the card-punching devils intended?
Sure - I use GUIs most of the time despite maintaining Ratatui. I much prefer them for coding agents as the amount of interactions and various things I want out of the interface are not handled by TUI well. GUIs are a vastly superior interface (I worked on the Codex TUI at OpenAI btw).
How much RAM does this use?
Nothing really - Ghostty has a bit of padding that's unused per cell - 16 bits. The implementation stores an ID into a lookup table with border combinations. If all of those were used and never cleaned up, 65535 at around 100 for each combination used, order of magnitude 25MB absolute worst case assuming no real optimization in place for this.
Having markdown tables render with zero-cell spacing (use underlines for horizontal and different background shades for vertical) is just one of the billion reasons I will never, ever, ever go back to closed-source coding agents.
This is gorgeous and I need it in trunk for both kitty and ghostty
Why do you people have to ruin terminals? If you wanted a web browser, we already have those, and they already have borders.
What if, dare I say it, we just served HTML over ssh, and let the browser render a proper UI, instead of layering hacks on top of hacks?
I think the other side of things is why have the graphic program run over ssh, just have raw text cli tools to manipulate what ever then have a program driving it. This for me is where emacs excels and some programs already fit well with this model and the fantastic (albeit hacky) tramp.
No, i really don't want to have electron level bloat as a terminal
Honestly why? It is not designed at all for that.
In fact it just makes it less secure.
You mean serving html over SSH?
It can already port forward so you can basically already do this, I don’t think that’s insecure per se.
Doing it through port forwarding is kinda bad though because generally access to network ports is not very well access controlled. Doing it through unix sockets (like x) would be much better but annoyingly browsers do not support http over uds
https://bugzilla.mozilla.org/show_bug.cgi?id=1688774 etc
Ssh lets you forward arbitrary ports, and forward X11 directly. Why is HTML more of a problem?
> HTML
> instead of hacks on top of hacks
Afraid that’s a contradiction. In this role, HTML is a pile of hacks upon hacks that was not originally intended for UIs.
But, yeah, people are trying to reinvent GUIs. By layering hacks over TTYs. If only there’d be some protocol for widgets and events over a stream connection…
> In this role, HTML is a pile of hacks upon hacks that was not originally intended for UIs.
I get where you're going and am genuinely curious, what does a "better" system designed to do what HTML/JS have become actually look like? (or what would one look like?) There has to be markup of some sort, and there has to be scriptable interactions of some sort, right? If not, how else would it work?
While we're figuring it out let's call this hypothetical protocol X.
We could even use `ssh -X` to send it over SSH if that flag is otherwise unused.
I think it would make sense if it had a client-server architecture.
And for the ecosystem's sake, punish (harshly) GUI toolkit developers that use bitmap rendering (GTK4 and Qt Quick applications are slow on ssh -X).
the thing with X and wayland etc is, that they are low level. which is fine if you want full control over the presentation. what HTML was trying to do is to go higher level, send over some data and let the browser figure out how to present it. but it went wrong and people are abusing it as a low level presentation interface like X or wayland.
what we are missing is a really high level protocol/system that takes really high level data and figures out the presentation on its own.
take a list of files for example. you can run ls -l in a terminal and it will attempt a crude text based formatting of the output. or you can run a full gui filemanager.
what if there was a command that would produce ls output as a table data structure, run in a sort of terminal that knows how to format a table nicely without the ls command having to include any formatting instructions? this terminal would have support for all sorts of data types and structures to display all of them nicely.
> Take a list of files for example. you can run ls -l in a terminal and it will attempt a crude text based formatting of the output. or you can run a full gui filemanager.
ls's output (and a few other programs') are human facing. They're not meant to be machine readable.
> what if there was a command that would produce ls output as a table data structure, run in a sort of terminal that knows how to format a table nicely without the ls command having to include any formatting instructions? this terminal would have support for all sorts of data types and structures to display all of them nicely.
Because that's just complexity for its own sake. What's simpler is to just create a shell alias for a particular view you want. The shell is for interactive usage, with some capabilities for automation. If something better is needed, you write a real program for it.
> what if there was a command that would produce ls output as a table data structure, run in a sort of terminal that knows how to format a table nicely without the ls command having to include any formatting instructions? this terminal would have support for all sorts of data types and structures to display all of them nicely.
So, powershell? For some reason only IT jockeys on Microsoft's death grip remember and use its full capability after learning it.
There are many problems with powershell, but to me the absolute worst, won't even touch it with a 10-feet pole, is UTF-16.
plan 9?
UI in a browser is a massive hack in itself, a clean design would be better. For example a compact semantic remote access protocol to use with UIs, something like SPICE but better optimized for networks, fully presentation-independent, and less janky. This would let everyone use a GUI framework of their choice with an adapter. But people building terminal GUIs are culturally not ready for this, they'd rather continue reinventing the wheel and piling up hacks in terminals. I wonder when they realize that proportional fonts are more readable when vertical alignment is handled separately, and introduce terminals with variable-width fonts. And the presentation layer (I suggest calling it CSS).
“….that don’t cost a cell.” How much do I hate AI hustle bro passive riddle writing.
A pet peeve of mine is that most people / WM get "useless gaps" wrong. You do want a gap between two terminals because otherwise there are times where things can get really confusing (where you don't know what belong to which terminal / window).
So the "borderless" (0 pixels) / zero "useless gaps" simply are confusing and... Wrong.
But then people who do use useless gap, as in TFA, get it wrong too: they typically add gaps near the monitor's edges.
Yeah. But no. Sorry but my monitor's edge is a perfectly fine gap. I don't risk mistaking it for anything.
So I modified my WM (the Awesome WM) to do proper "useless gaps" computation and remove the ones that really are useless: and that'd be those near the monitor's physical border.
I mean: nearly everybody gets it wrong, but it's not complicated to get it right, is it?
P.S: even the terminology is wrong... People calling those "useless gaps" haven't got a clue: they are useful. The only useless ones are the one on the monitor's edge (if your monitor has physical edges, which it probably does).
Man, I’d really like a monitor with no edges. One that just keeps going and going, past my window and the tree outside until it’s completely out of sight. 100% useful area.
“That don’t cost a cell.” Can’t stand AI bro speak.
One benefit worth calling out: the box-drawing fallback is worse than it looks for CJK users. U+2500-257F and the half/full block elements are East Asian "Ambiguous" in UAX #11, so in a terminal configured to render ambiguous as double width (xterm's cjkWidth, iTerm2's checkbox) every line glyph takes two cells and the whole frame lands in the wrong place. Cell borders aren't text, so that failure class just disappears.
The part I'd want specced early: borders are per-cell, but cells aren't per-grapheme. A CJK char owns two cells plus a continuation placeholder, and combining marks or ZWJ emoji sequences own zero-or-more depending on whose wcwidth table you ask, so app and terminal routinely disagree across Unicode versions. Today that shows up as a vertical bar one column off; with cell borders it becomes a line that stops short of the corner. Does the border attach to the cell or to the grapheme cluster?
> If that bothers you, feel free to close the tab.
I probably missed on something, but I stopped there