Hashimoto is an absolute wizard, but what I find most compelling about him is his absolutely uncanny ability to segment and abstract systems and interfaces in a way for maximum composability and minimal entanglement. He's like the walking embodiment of Rich Hickey's Simple Made Easy philosophy. It's like he designs software systems in such a way that they have no choice but to operate correctly and predictably.
Also I just tried Ghostty for the first time. With iTerm2 and the Zsh/Powerlevel10k theme, there's an extremely brief but perceptible lag from running a command and the render. In ghostty it feels actually instant.
Because maintaining is not as fun as starting? Because other people can maintain a well structured project but not as many people can start something from scratch? Because his skills can be better used elsewhere?
That sounds like the Google philosophy though, where smart people come up with ideas and write the initial implementation (then get promoted) and other less smart people take over, and you end up with a mess that had great potential like Bazel?
My understanding is that this is not a philosophy at all, it is an incidental result of a bad set of misaligned incentives, where to be promoted you need to start something new or drive growth and post launch maintenance/growth is weighted much less
That sounds like what Drew DeVault does (did?). He made sway, aerc, and others but largely doesn't work on them now AFAIK, passing off work to other people such as emersion.
To add... The missing scrollbars feel painful to me. I don't even want them for scrolling, really. I want the visual reference for:
"How much content is in this window"
and
"Where am I in that content"
I still use it daily but it means I have to switch tools for certain things, and reading log files or log output is one of the more common reasons I switch.
I should probably look into trying to get the scrollback info into my statusline, No idea how easy or hard that is - so if someone has done it, feel free to shoot me pointers.
Centralized logging, a la Splunk is the gold (with what they charge, probably platinum) standard here. Watching someone who knows what they're doing at a Splunk console for a distributed system they understand is a thing of beauty. It's just a shame they charge so much.
Same here, scroll bars are literally the one missing feature keeping me from making Ghostty my daily driver. But the release notes roadmap mentions they will add them for 1.3. Waiting with anticipation.
Maybe I'm misunderstanding your request but do you not pipe outputs into a pager? (Less, more, batcat, etc) I feel like I'm probably not understanding the context though.
I'm still not quite sure what you mean. You can still {|,>} output from a program, so less and grep still help.
Or do you mean when you inevitably forget? Well then yeah you're at the mercy of your terminal but as others mentioned ghostty has a hack to help as well as some other terminals. But this should also help reinforce why you should pipe more often and write to files (or tee). It sucks but not making the same mistakes in the future and learning better habits will help you write better code and use better practices.
But that's the age old problem of "you can't analyze the data you didn't record" and that's a footgun you'll experience in every programming language, every experiment, and across many parts of life. Better to record and throw it away than not record and regret it.
It IS recorded. It's right there in scrollback (Literally the default buffer to record). It's easily accessible with most tooling, including nice scrolling, mouse support, find/search, etc...
Except in Ghostty, it's not so accessible. No find, no scrollbars.
Although deciding when to do that would be easier if I had a better indicator for just how much scrollback content exists. Ex - if it's 3 pages... I'll just scroll through it. If it's 3000 pages... time to dump to file.
So no - by default I use a pager... just about never. Why would I when I have absolutely everything in scrollback by default 99.9% of the time?
---
Don't confuse your preferences with "correct" :P
I'm doing just fine with code and best practices, I'm simply stating that this is a rough edge on an otherwise lovely tool.
In the latest version Shift + ctrl + alt + j copies the entire screen and opens it in what I assume is $EDITOR (maybe this can be configured). Or just shift + ctrl + j to copy the screen to temp file and pate the path. Either way, it then just becomes configuring it to use a pager by default, or `shift+ctrl+j` and then | less. This is similar to how kitty does search, by essentially outsourcing it to an external pager. I was put off by it at first, but it's very much in the UNIX philosophy.
I find the search through a file actually makes sense.
If you don't know, you can send the log to a file, and open that file to look through it. More powerful than just a search next, as you can have instance counts, search with regexp and all the bells and whistles and it virtually stops the logging.
At first I thought the same as you, now I've become quite partial to this concept. I hope they don't remove this.
There's a few things like this but I'll give a pass because how new it is. Were this a mature project I'd be upset at incompetence, but given how young it is I think it's too early to be harsh.
Plus, they're very open about what they're doing and prioritizing. As another commenter said, it's coming soon. For the rest, open feature requests, you might have needs that others didn't think of or even realize they needed
I honestly didn’t even realize it was missing until I was reading this thread, cause I’m always in a zellij session which has a “search in default editor” key.
Pop the whole scrollback in helix, where I can search, select, jump around, paste stuff into a scratch buffer. It’s slick. It’s got a normal search too. But yeah I haven’t used a built in emulator search in a while I guess!
The nerds would do sooo much better for ourselves if we could build a momentum behind comments like this.
Finding mechanistic (& programmatic) sympathy counts for so so much! Shapes the arch of software so much! But it's such invisible unknown work to most people, not so overtly clear & obvious but something that constantly builds day after day, person after person, incremental 0.2% gains compounded by lack of friction.
As well as just promoting good practioners, it feels like discourse about software architecture has really fallen off. We are deep inside rabbit holes specific to this framework or that, and there's only occasional popping out to free air to bring back some observations from the burrows. Ideally we'd have many more volumes of Architecture of Open Source Applications (2011), for example, to really dive into what is, to give us some common referents to learn from and talk about. https://aosabook.org/en/
This is all so core, so worth getting deep on & looking how things are assembled, what the interfaces and modules and shapes look like, what the tradeoffs were. But it remains chiefly an arcane art, one that most developers much less most businesses haven't developed a refinement or taste for.
I just want to say, what a dream. To have wealth and be able to create projects for the sake of going the projects. To not have to make the concessions of quality for profits.
There's an old Knuth quote:
> In fact what I would like to see is thousands of computer scientists let loose to do whatever they want. That's what really advances the field.
And I think we're seeing more and more that these projects made with love are successful. That without the hyper fixation on money we can build good projects that make big changes in a world.
In some sense I'm a bit envious of Mitchell but truthfully these types of things make me more question how we've constructed our society and economy. It shouldn't require one to start with wealth to be able to build things that have such an impact. What needs to be changed where we can live up to what Knuth proclaimed. I'm sure all of us have had experiences where were we given the time (and usually not much) we could make things so much better. But we make many sacrifices when we rush. Which leads to more good advice by Knuth
> If you optimize everything, you will always be unhappy.
At what point do we push back? We see that the people we really look up to did things so differently. Knuth himself expressed how detail obsessed he was, and such a claim is common among the grey breads.
Of course, things change, but are we creating a world with no wizards? Are we creating a world where we reward people for solving problems and making our lives easier? Or are we just maximizing some score of a pointless game?
I'd love to live in a world with a thousand more Mitchells, following their passions without the burden of needing to justify decisions to a board who has no interest in quality. How do we create that world?
Like you said, what a legend. But, how do we make more legends?
What we are missing is a sponsorship model that is able to fun this kind of thing. It's not infeasible - look at university grants dispersement as an example of the overwrought predecessor of this. A ton of money moves through such institutions, and funds a lot of interesting projects.
One thing I've been envisioning is something like a "certified B corporation" style qualification that companies can get that indicates they contribute financially back to open source commensurate to the amount of it they consume to run their core business. If everything you do runs through open source software, in a moral sense, one can make the argument that you owe something back to it.
Having recently come out of grad school, tell me about it...
But the same general problem exists in industry. Our fear of doing things non-optimally only results in a less optimal solution. It's a risky move to take no risks.
In both academia and industry you see the same people doing the same things in the same way. It's no wonder things don't change. You can't have a paradigm shift by following the paradigm (playing it safe). I feel like this is a big shame in both tech and academia as the histories of these have always been made by those who rocked the boat. At some point we just have to admit we're not very good at predicting the future and instead of trying to predict what will be the most successful we should fund passion. I'm sure charlatans will get funded too, but its not like we're doing a good job at preventing that from happening now anyways...
Also, in real life, he's a genuine nice guy. I had lunch with him and Armon at a conference when they worked at Kiip, before they founded Hashicorp. Mitchell (to me) was the proper hacker. He just loved everything about computing. This particular conference was about distributed systems and he was just geeking out on everything.
I walked away thinking that no matter what they did, they'd probably be successful. I was extremely happy to find Ghostty and have been using it ever since.
I had tried it briefly previously but in the last couple months I think I have made the permanent switch from iterm2. It's so much snappier and simpler and also reliably handles text reflowing when a long line wraps, which was a constant problem I had in iterm2, where it would insert fake newlines when copying out text that was wrapped.
I also like that I can have my config in a little plaintext file and just drop it onto a new computer and get the same keybindings. I am using the terminator keybindings for creating and navigating between split panes.
Basically MariaDB. He (original creator of MySQL) sold MySQL for a Billion dollars (iirc) forked it, created MariaDB and kept at it. It's been what 15 years now?
You made me look into HashiCorp again. It's a great company from a developers perspective, in terms of the many useful tools created: Vagrant, Packer, Nomad, Consul, etc
Actually, that's the question (or rather the feeling) I had even before seeing these numbers. Just by reading the docs and looking at everything they built (and how they built it) made me wonder if they spend more than they make. That's a really funny feeling I never had before. Like, surely economies can't be that forgiving when it comes to polishing things. Is that what's happening to the company? Mere overspending?
This is super cool, it will be so nice to have a truly omni-platform terminal emulator that can even conceivably extend to iOS and Android.
Aside: I didn't realize Ghostty was written in Zig, wow. The first Zig-thing I'm aware of using on a regular basis. It's amusing the repository structure looks exactly like a Golang layout, haha.
Since go 1.25 you can declare the what is the “root” of a go module, which is a good thing. Not everyone wants the top level directory filled with go files specially in a multi language repo. So this idea that /src is not good go directory will soon be outdated.
The passion and attention to detail Mitchell has for developer experience is beyond me. I remember the first time I used Vagrant (2011, Santa Monica California) like it was yesterday it was that impactful.
I never thought in a million years I would even think of ditching iTerm2 but when Ghostty dropped I installed it and fell in love.
Great question. For me it feels faster (anecdotally) and certain TUI components don’t break. It also is probably my bias of loving his work as well as everyone at work using it. Hive mentality I suppose.
My experience is similar. Stuff doesn't break and it feels extremely snappy. I've switched to Zed for the same reason. I came to realize that I work much better with faster, simpler software, even if it's missing some nice-to-haves. It feels like you're closer to the problem you're trying to solve.
- No way (that I know of) to select previous output or specific string and copy with only keyboard shortcuts.
- No ⌘. sending CTRL-C (muscle memory and being advertised as native to the Mac is what one would expect).
- Fonts still don't render as nicely as in Terminal.app. I've fiddled with `font-thicken-strength` and it's close, but not quite. Probably impossible or very hard to replicate due to its Metal rendering nature, but when all you do is look at text all day, it matters.
I've been delaying a major migration to NeoVim on [modern-terminal-I-can-never-decide-which] for years.
Wezterm, Ghostty, iTerm2. None is exactly perfect, so I just keep watching them develop.
I recently migrated from vim to neovim and you can just migrate everything. I forgot where I found this but put this in ~/.config/nvim/init.vim
" Load vim configs
set runtimepath^=~/.vim runtimepath+=~/.vim/after
let &packpath = &runtimepath
source ~/.vimrc
Then in my zshrc (well... I organize differently) I have the function
function _exists() {
command -v "$1" &> /dev/null
}
alias_vim() {
if (_exists nvim)
then
alias vi='nvim'
alias vim='nvim'
elif (_exists vim)
then
alias vi='vim'
fi
}
FWIW, zsh has a commands hash to make stuff like this potentially easier and cleaner. The following isn't quite how I'd do it, but is functionally equivalent.
(( $+commands[vim] )) && alias vi=vim
(( $+commands[nvim] )) && alias vi{,m}=nvim
I wasn't aware of +commands, thanks. Though I'm not sure this is easier and you can translate mine to bash trivially
Yours isn't technically equivalent though it is functionally. If we have both vim and nvim then we'll alias vi twice.
Also, your program provides less clarity. It saves lines but at a large cost to readability. I tend to share my dotfiles with newbies a lot so readability is very meaningful.
I wrap mine up in functions too because at the end of my alias file I can add something like this
It's a bit overkill and I never expect a simple alias like that to fail but there are three distinct advantages I get for being just a tad more verbose:
1) I have more complicated versions to deal with things like fd and batcat which have different names different operating systems (`fd` vs `fd-find` / `bat` vs `batcat`) and some additional configurations.
2) I can disable the alias by commenting out one line
3) knowing exactly where the alias is being loaded and thus what aliases are loaded.
Bonus) fails loudly but continues (it's an alias, I don't want you fail fail)
A few extra keystrokes are worth this advantage imo. Especially since I'm using vim so it's actually not any additional typing lol
It's style, so the choice is up to you and they'll accomplish the same things, but I'm just explaining why I use this design pattern. I've only given a stripped down snippet of code so I hope this context helps explain the larger pictures and how it can be used for larger needs.
(Surely this issue must've been discussed/debated elsewhere ad nauseum because it seems an odd design decision to leave out such a common macos binding...)
But having only used ghostty as-is and getting bamboozled by the copy paste situation, this is game changing. I was just going to wait till preferences had a GUI/TUI.. So thanks!
> - No ⌘. sending CTRL-C (muscle memory and being advertised as native to the Mac is what one would expect).
For what it's worth, that shortcut also doesn't work in the Jetbrains terminal, Alacritty, Kitty, and even iTerm2. The only terminal emulator I've tried so far where that works is the built-in Terminal.app.
Is there a blog post anywhere that explains why it has taken so long to get search working? Ghostty is such a nice app but that's such a fundamental feature that there has to be a good reason... It's the only thing keeping me on iTerm2.
I don't want to presume your use case, but Ghostty has a command for dumping the buffer to a file, which I use for processing output "too late" to use grep.
Try a pager instead. Batcat is more feature rich but there's always the good old less (and more) command. Both work great with grep. I do things like the following multiple times a day
$ cat foo.log | less
$ cat foo.log | $PAGER
$ cat foo.log | grep 09-23-2025 | less
Side note / pro tip: on a new line in the terminal press control-x control-e. If you're in zsh you need to edit your config but this will work out of the box for bash.
I do. Setting up log files for each and every command seems tedious. I would rather just cmd+f to search and have it work for everything, as it is a property of the terminal, rather than a specific command.
By the rollercoaster of points this post is experiencing, I may have expressed myself poorly.
To be clear, the concept of a modern, cross-platform Terminal core with native GUI implementations for each major vendor is amazing and I'm very glad it exists. It's just not quite there yet for me and I don't know if it ever will be. But I'm very excited by it and hopeful.
Interesting, I really prefer the font rendering in Ghostty! I'm with you on the first two items, though.
Despite these shortcomings I still use it daily since the first time I tried it, and have come to really love the project in general. I often won't hop to new software if it offers friction like this, but Ghostty pulled me in despite some snags.
Zed is also wonderful software, though it took me way longer to commit to using it daily.
- It doesn’t respect the user’s configured fonts.
- Clipboard support is inconsistent.
- The GTK4 UI is quirky, and has silly slow animations in situations which don’t make sense.
The first two sound like more serious issues, and I’m surprised that others haven’t found them dealbreakers.
I want to like it too. I even joined the Discord I was so excited about it. Unfortunately even at version 1.2.0, which I understand is a total rewrite on Linux, it's only usable for minutes before this bug kicks in https://github.com/ghostty-org/ghostty/issues/2210
It means cancel, stop. You can use it to stop progress bars, dismiss dialog boxes, terminate some well-behaved spinning beach balls, cancel a drag before dropping, etc. It's been on the platform since the classic Mac OS days.
> (This is somehow fixed in 1.2 but it requires additional entry in the config for whatever reason. Can't it just work?)
"For whatever reason": because it requires modifying `ssh` and doing that by default is a really sketchy thing to do because it is a very security sensitive tool. No program (Ghostty included) should be mucking with that by default. We want users to be aware.
"Can't it just work?": No, because the entire mechanism is flawed, and I didn't invent the mechanism. It's like asking, "why must I drive to my destination, can't I just teleport?" The entire premise of the question is silly, and it's not the car's fault (Ghostty is the car in this example). For those who want to learn more about "the mechanism": read Ghostty's terminfo page, but also just do some light web searching on how terminfo works. Its a total nightmare.
I really wish that weren't the case, I really do. I promise its just as annoying to me as a user and more annoying to me as a maintainer to have to hear people complain about this repeatedly when I'm not doing anything wrong, personally. I'm playing by the rules. The rules are just bullshit.
We'll continue to make enhancements to improve this while we wait for Ghostty's terminfo to propagate to every machine in the world. It will, it'll just take... a long time. Next up we plan on introducing a `ghostty +ssh` command that you can drop-in replace most `ssh` usage with and it'll automagically just work.
Ouch, I did not expect my complaining would get a response from the author!
Just to be clear, I do think Ghostty is amazing piece of software and it's so fast that it's hard to believe, so thank you for the hard work.
> read Ghostty's terminfo page, but also just do some light web searching on how terminfo works. Its a total nightmare.
Oh, yea, I totally get that. I actually _did_ try to understand what the issue is, but gave up on that time-sink rabbit hole and decided I might give it another try once the xterm-ghostty is more popular.
I use ghostty daily now. Switched recently. In macOS I can re-map now caps lock to be cmd and it works for cmd+c. Everything else is great, sensible defaults and what's not easily customizable.
Gruvbox light theme looks great too.
The fact it's written in Zig is awesome too, if you ever question if Zig is ready, ghostty is your answer to that.
Not seeing myself going back. It's great experience.
Tip: if you combine your ghostty flow with aerospace, it's nearly perfect setup for your keyboard only experience on mac.
I use terminology today, which is also "fast." The major issue I run into with ghostty vs terminology is that ssh in ghostty is broken due to the terminfo thing. I don't fully understand why ghostty can't pretend to be a different terminal that remotes know about (for now), but I assume there's some reason.
Have you tried the suggestions in https://ghostty.org/docs/help/terminfo#ssh? I don't know what issue you may be experiencing but this solved my issue with using htop in an ssh session.
That is essentially the 2nd solution recommended. I think ghostty could do this itself as a hack. If the target binary is ssh (heuristic, but, ok), inject TERM=xterm-256color into env of the child process at launch.
I guess it isn't a huge deal to have every user to modify their ssh_config instead, but it's an ergonomic pain point for many new users.
Fast. Very fast. Good configuration and theming options too. Literally no lag, even with thousands of lines scrollback. It doesn't get in your way on your day to day terminal activities. Feels natural.
Everyone telling you that it's fast because iTerm2 is slow. Terminal.app is already faster than iTerm2 and on par o very close to most other alternatives in terms of speed.
I also used Terminal until recently and don't use any of the advanced features alternatives provide. The main reason to switch from Terminal.app is truecolor support. The terminfo thing is annoying but I just setenv TERM in ssh config. Better split panes is nice. Configuration in a text file is a matter of taste, but documentation is good.
I think you’re thinking of TigerBeetle, not Clickhouse, which is a quite performant db with a fascinating simulation-tested story and proof of performance / safety.
Several years ago I was crazy frustrated with how slow 'less' is for some things (regex search and handling long lines) and was slowing down my day to day operations work. I very briefly entertained an idea to write my own pager to fix those problems but it would have been way too complex to do in my spare time. If libghostty existed then it might have been a different story. If libghostty is done well I think it could be an enabler for many types of terminal applications. I'm excited to see where it goes.
I've been keeping an eye on this in the hopes that text-reflow (edit: including for scrollback) could be solved in Neovim-based terminals [1]. I'm loving the innovation Ghostty is bringing to the terminal space.
Do you consider yourself a neovim terminal power user?
I tried a while back to invert my workflow (from tmux driving neovim to neovim driving terminals) because I thought it might be easier to only ever have one buffer open for a given file, instead of attempting to open a file in a given pane only to realize that it's already open in a different neovim instance in a different pane.
When I was testing that stuff out I don't think I noticed particular issues with text reflow that would benefit from being solved by swapping to libghostty, rather my pain points were just about how to adjust to the different paradigm. I'd be curious to hear more about someone who is all in on Neovim embedded terminals (and possibly how libghostty might make it better).
I'm all in on Neovim terminals, having a remote development setup means it keeps my terminal with my neovim window (I use nvim-qt).
Also not sure how ghostty would help, haven't noticed text reflowing issues.
It's not bad, a little awkward getting used to:
- you might want a plugin to give you a "persistent" terminal across all tabs
- I still haven't found a way to clear scroll back while a command is running
- I had to set up mappings for easier exiting terminal mode (c-\ c-n really sucks)
- I had to set up events so whenever a terminal buffer is focused it immediately enters insert mode. While I love vim, I've never wanted modal editing in a terminal
I do indeed live in the terminal (all day due to work), but tmux adds too much value for me to do all terminal management in Neovim (tmux session-management being what I use most). I've just encountered too many visual "glitches" in the Neovim terminal to rely on it for everything. That's not to say, however, that I never use the built-in Neovim :terminal.
> I thought it might be easier to only ever have one buffer open for a given file, instead of attempting to open a file in a given pane only to realize that it's already open in a different neovim instance
I'd be curious to hear more about how tmux helps you — I tried it and besides keeping a permanent session open on a remote server to me I didn't find much use for it compared to regular terminal tabs
I use it daily locally, and find it amusing how many only think of it as being useful on remote servers (not to invalidate your use-case -- I'm just contrasting my own use). As a precursor, I view UNIX as my IDE, of which tmux is a part: this IDE runs on Windows (WSL2), macOS, Linux, and Android (Termux). That aside, here are a few reasons I find tmux to be useful in this concoction of tools:
- Session management. I've written custom scripts for myself around this (zoxide + fzf). If you want to see how this can be used, look at ThePrimagen's workflow. I don't use his scripts but he has a good demo of how he harnesses sessions.
- Unified scrollback management - easily search the scrollback, yank it, etc. My favorite thing to do is to yank part of the scrollback, then `Prefix+B,=` to list everything I've yanked (think of this like a "clipboard manager" specific to tmux), select an entry, and press `e` to edit it in `$EDITOR`.
- This one might be a stretch, but I tend to try and use only terminal tools (without being utterly insane) because then tmux can be my "tiling window manager" no matter what OS I'm on. Oh, I have to use Windows for work? Not to worry, tmux runs in WSL2, as do most of my preferred tools, so I feel mostly at home even though I normally really dislike Windows.
- It's scriptable. Read `man tmux` and use your imagination!
Notwithstanding any of that, there are cons, the most apparent one being that I am limited to text-based tools this way. An example of this: getting images to work in tmux, though many modern terminal emulators support them, is a huge pane, so I haven't bothered.
I think my problem is when I realize that I had unsaved changes open in a different neovim instance. If the file was not dirty in any other open neovim instances then I don't have the same problem.
I use ghostty on (my?) air mac that my brother gifted to me which was his previous device for coding / general purpose etc.
Ghostty is really good and I genuinely appreciate this.
Honestly, Might seem like a small thing but the fact that this doesn't even depend on libc seems to be a really great thing to me and I kinda cherish it!
I have heard this from a lot of people, yet here I am, using a terminal that supports this, and have yet to ever do it.
Can you help me understand what workflows you depend on Ctrl+f for? I wonder if I am missing something big.
Don't you ever need to search through a program's output; e.g., to find what failed? Otherwise you'll have to remember to tee everything to a file every time you run a command.
You know, I can't remember ever doing this in a terminal. Not because it doesn't sound useful (it sounds VERY useful), it just sort-of never occurred to me as a thing you could do. It's just muscle memory at this point to rerun the command with a pipe to a pager and search there.
I get ya, but this would be more of my case if my terminal didn't have scrollback search available (or if I really wanted to scrutinize the output carefully).
In foot I just ctrl+shift+r and search back the references one-by-one in a mode, which I guess is 90% of my use case for scrollback.
I guess that's what I always do. or grep or awk or sed.. Maybe if a program takes hours to return it's helpful to be able to search the terminal, but otherwise i'm just pushing up with my arrow and `| grep -v UGLY | grep BEAUTIFUL`
It could take a long time to run, and my time is not free. The native MacOS terminal has a buffer search, why doesn't ghostty? It is a rudimentary feature.
You can map something like cmd+shift+f to open the entire buffer in your default text editor, this has been sufficient for a lot of folks - myself included - while we wait for native scrollback search to land.
I started using it months ago and don't miss that so much, if I really can't rerun the command and pipe it to less/grep. I'll just use the mouse to copy it to the clipboard and then run `xclip -o` (linux) or `pbpaste` (mac) and pipe it to grep. I know there's a keybinding for putting the buffer in a file, but copying with the mouse is faster than finding it.
Also, not sure if this is by default or it picked it up from my old iTerm2 configuration, but cmd+shift+up/down navigates through prompt lines so it's easy to find the start of a long command. My PS1 in zsh is:
As someone who lives in the terminal, I can't say I've ever had the need to do that. It's only by reading the comments that I've realised that there's no search in Ghostty.
> All of this is packaged up into a single zero-dependency C API (it doesn't even rely on libc), allowing it to be easily embedded into any popular language ecosystem.
@mitchellh That's super exciting. I'm wondering how easy it would be to integrate in the debugger frontend I'm working on. Do you already know what the C API will look like?
I want to like Ghosty, but not being able to search has been a deal breaker. Guess the alternatives to search are use tmux or grep the ghosty log, but tmux and I have never clicked, and searching the log is not user friendly for how often I find myself wanting to search terminal history. Ghosty has a lot of features I would want, but somehow it is missing search.
https://www.reddit.com/r/Ghostty/comments/1i0g2ge/how_to_fin...
Maybe this concern doesn't apply to libghostty, as I think search would be more of a UI concern.
Serial port or serial port over "ethernet"[1] (technically it's 8p8c[2]).
My router doesn't have a video port, but it does have a dummy terminal port. I had to scrounge a video card for my server to set it up, but it does have a serial port [3]. So that would have been nice.
Also would be nice for a modern remake of dumb terminal with abandoned monitors.
I'm really digging Ghosty's shaders. It finally feels like a gimmicky terminal done right. Want a Fallout or CRT theme without compromising on other features? You can finally have it. It brings back memories of the early days of Linux when everything was fun and not trying to mimic the most bland OS out there. Think Enlightenment before everything got gnomified.
I want to use this to modernize vterm in Emacs. If I could only synchronize the terminal cursor and the Emacs point, and preserve lines as lines, not split them...
vterm flickers a lot with any busier TUI. Claude Code sucks in vterm. I'd also really look forward to using libghostty to try an alternative to vterm in emacs.
I can’t figure out how to get the visidata frequency analysis histograms to render properly. Some render as blocks as expected but others are those diamond-question marks. I know it’s a silly hang up but it’s what’s keeping me on iterm. It’s seemingly impossible to search for (or at least I haven’t figured out the right keywords) to troubleshoot too.
Ghostty has the capability to parse Tmux control mode (and this blog post is about libghostty-vt). Ghostty the GUI is missing the capability to map that to GUI elements. But Ghostty understands Tmux control mode today: https://github.com/ghostty-org/ghostty/blob/main/src/termina...
I love the idea, and I'm doing something similar in Ruby (which, yes, means it will be far less universal than a compiled library; that's an intentional tradeoff - I don't need it to be universal, and even less so when someone is willing to invest far more effort into something that is universal) inspired by AmigaOS, where a terminal window was easily accessible to any application whether it was started from the terminal or not, by opening a file handle to "CON:" (which meant e.g. scripts could also pipe this directly into a newly opened terminal window).
It's one of a few "lost capabilities" of AmigaOS I'm glad to see getting addressed. You can of course fix this today by using one of a variety of terminal widgets, but the ease of being able to just spin up a full terminal window from "anywhere" was very powerful for simple utilities etc.
Really looking forward to this, writing my own personal editor in Ruby and not having a library for parsing terminal input is leading me to write my own kitty keyboard protocol parser (which ghostty implements) and having to hand-code lookup tables is driving me batty. I really don't like the idea of using an existing TUI framework as none of them actually implement the kitty protocol.
Have you released the project? I am somewhat curious!
Also I just went from a somewhat (niche?) languages rabbit hole and I refound crystal / crystal is in my mind currently..
So what are your thoughts on using something like crystal instead of ruby because they are similar in what I know and crystal can create a static binary while being (faster?) than ruby
I like crystal but without pure OO and the fact that the community is so far behind Go or even Scala it's hard for me to see a good use for it.
You can create static binaries with ruby, just use mruby. I had a bunch of small scripts in crystal that I replaced with mruby and never noticed the difference. You don't get runtime reflection or ruby's enormous library base, but it could fill the need. I may end up transcribing my editor, or perhaps just the parser, to mruby once it's ready.
If mruby can make static ruby then that is actually kinda great, I didn't knew that.
I was thinking more of in terms of performance for some similar developer experience
Pardon me but isn't crystal OOP? i feel like it was because I was introduced to crystal lang via some youtuber some long time ago and the comments said that they didn't use because it was OOP
Also if mruby is so great (which it sounds here), then surely most ruby contributers can migrate over to mruby if that has these gains. I think mruby compiles to C as the intermediate right? So there are definitely some limitations as compared to being a interpreted langauge but that can be somewhat mitigated, no?
I am not a ruby expert or a crystal expert. Because If I was, I would be telling you your future (a slight pun ofc of being a future teller but yes I don't know much about it except the puts syntax or that 100 seconds video from fireship which gives me 10 year experience in ruby :p )
Crystal is object-oriented, but isn't a pure object system. In a pure object system, literally everything is, or can be turned into an object. Smalltalk is the only other example of a language family that offers a pure object system.
With a pure object system everything is accessible by the system at runtime. It's ruby's biggest strength, but also the superpowers it grants aren't always used for good, which is why a lot of coders hate working in it. I only code for myself, my coding career is behind me, so I don't have to worry about other people's code.
Mruby's big downside is precisely that it's compiled, so the library base available is very small, and there's not much point in building on it, though a few projects have, such as DragonRuby. It also doesn't have a pure object system. Ruby has thousands of gems, mruby dozens, and you can see them all on this page: https://mruby.org/libraries/
It is Ghostty that makes me believe Zig is ready for industry, and I also want to make a game engine targeting Android and iOS first, however, JNI is really annoying and I hate Google Android. I have tried Bevy for Android and as you can see it's not the fault of game engine but Google. So, use Unity or challenge it.
Just do it, thanks Gostty!
The core issue is that it's allocating a new tmux pane with the sole goal of mirroring line numbers; it would be nice if they synced up in the same pane, avoiding the above issue.
Piping it into neovim is an option that you can do on both neovim and zellij. zellij loses colors, and neovim is probably the best solution to this problem but then again I don't want to have to remember to turn on/off line numbers every time and I personally like one-off panes. Separation of responsibilities, I guess.
Long-winded rant to basically say: would a standard like this solve my issue easier? From what I understand of terminals, I would need to parse the underlying pty, maintain a scrollback buffer internally in the wrapper shim, and also be able to dynamically adjust toggling line numbers on/off.
If I'm doing this kind of translation, how "leaky" will the abstraction be until I'm basically re-implementing the logic in my middle layer, assuming that "for free" I can get the translation both in and out from the pty?
I've been trying to look closer at TUI tools, but that's what really bothered me. Given just how god awful the VT protocol is, you could get the state machine parsing correct, but the developer still has to learn basically every little quirk that was added over the years, no?
(And before someone makes a false equivalence, no, this isn't the case even with languages like c++ - I'm still learning quirks about it to this day, but I don't have to learn the entire thing to build proper, robust code. It does not seem the same with something like the VT protocol. So yes, I'm aware that some learning should take place, but I'm wondering how structured of a developer experience this will end up being.)
Someone call @keithw. He put a tremendous amount of brainpower into getting the Unicode side of this puzzle "right" (or, as "right" as it is possible to be) when he wrote mosh. I'm sure he and Hashimoto could have a grand old (but zero-width non-joining) conversation!
Ghostty is amazing. The attention to detail means that the basics work right, which I couldn't say about any other "modern" terminal apps. I still wish for xterm-like text selection, but otherwise this is the best terminal out there, at least on macOS (Apple Terminal.app is second).
idk why i am the only person that dislikes ghostty, i just have a tendency to put my machines under heavy load and keep using the cli and ghostty chugs super super badly when there's any gpu or cpu load, especially compared to "slow" terminals. zero issues with iterm2 on macos or gnome terminal on linux under load but that plus zero bells or whistles makes ghostty seem like a purposeless toy to me.
libvterm is great. Ghostty supports many more features, but the most important I hear from other libvterm users are that it's missing scrollback and reflow on resize, which are both pretty major pieces of functionality.
Will there also be a React component? Or is this not intended for web frontends? I was wondering, because Vercel et al. were mentioned in the beginning.
I have a confession: I don't know the difference between a console, terminal (-emulator?!?), shell, bash and all these things. It's all just a black window with text for me. I've never understood why people talk so much about git shell when my normal windows cmd/Powershell can do just the same git commands.
I'm also a prolific and successful software engineer. These two things shouldn't mix. So please tell me what I missed in life when these things are completely alien to me.
> The terminal seems simple at first (you type in commands and run them! no big deal!), but the more you learn, the more you notice a million little inconsistencies (why does pressing the arrow keys sometimes print out "^[[D"? why does selecting text sometimes not work? why are the colours sometimes unreadable?) that make it feel like an inscrutable black box. And it often doesn’t feel worth it to learn more because documentation about the terminal is so fragmented and full of obscure jargon.
> But! Understanding just a little more about the terminal can make your experience WAY better. You can quickly recognize what’s causing a problem (“oh, my arrow keys aren’t working because this is one of those annoying REPLs that doesn’t support arrow keys!”) and immediately fix it (“I’ll use rlwrap!). Or you can turn “wow, this text is unreadable” into “oh, my terminal emulator is responsible for colours! I’ll just go into the settings and reconfigure my colours!”.
I recently ditched VS Code in favor of Neovim and thanks to Ghostty transition has been a success. As a Mac user I use the cmd key in tons of shortcuts and it just worked out of the box, no need to send weird escape sequences
Total newb here - but lets say i want to make a terminal based video game (think old school text / visuals / menus), will something like Libghostty be something relevant to that?
To expand on the other replies: There are two very different tasks here.
This library would be for the task of taking a received ECMA-48 character stream and turning that into a series of actions on some sort of model of a terminal's display, keyboard, and mouse.
You are looking at a library for the (roughly) inverse task of taking a model of UI widgets and turning that into a transmitted ECMA-48 character stream.
I imagine if you wanted to make a self-contained application so non-command-line users can play it, then maybe you could use it. Otherwise, probably out of scope of the game itself.
Mitchell and bellard and to somewhat degree fwsgonzo (you might not know him but I absolutely like his work on libriscv, the game he's working and the kvm sandbox which he's working on too) are my definitions of literal coding chads.
People should really measure coding proficiency as being 10% of these guys or some fraction instead of whatever metric they are using now (slightly kidding ofc)
They are the 10x developers and I feel like this sentiment is echoed all across the industry
Sqlite team also seems to be one of the 100x people imo. Maybe I just cherish sqlite's simplicity and its hyperfocus and it delievers exactly what it says (somewhat), a lot!
Definitely feel like we need a forum where we can all compile a list of 100x developers and vote who is the best 100x developer
I feel like although Mitchell is up in the top three. To me bellard/torvalds tops it all of if you consider it from a pure software perspective and not that Mitchell went back from a billion $ empire to hacking again
Bellard has created qemu and ffmpeg and tcc/jslinux/ I think even xterm js kind of came somehow due to bellard so a lot of terminal emulators and their recent work is just as fascinating.
Torvalds with linux and git are two back to back things which we know and love and use today.
Bellard is just a little less known because unlike linus torvalds who named the os as linux ie. after himself. Ffmpeg and qemu don't really tell them that they are created by the same guy.
The nix package for ghostty is broken/unavailable on MacOS at the moment, but installing the brew cask (`brew install --cask ghostty`) and a basic config file:
Got me almost everything I used from iTerm2. Nerdfont just worked (in iTerm I think you have to use the `Use a different font for non-ASCII text` setting)
... is the perfect example of why "align continuation lines to the open bracket" is the absolute worse option versus the simpler, consistent and only logical choice of "align continuation lines with an extra indentation", aka
Indeed. Also note this isn't the real public C API. This is, as I noted in the blog post as a disclaimer, an internal-only C API so it is admitedly very ugly.
(I assume you know this, just adding context for other readers)
Yep, sorry in my mind I was going to mention it (with "even if it's only me seeing private code like this" or something similar) and later forgot after some edits to fix the formatting...
Anyway, I hope it doesn't get lost that this comment was only meant to be half informative, half public statement, and half a light joke :)
I wish they had used something like this in Shopify, instead they decided it was a good idea to embed VScode in the browser in case users want to
make a small modification to theme files, it even has a 5 second boot time.
I don't think its really practical to do so. wasm can't directly access the canvas or webgpu contexts. it has to route date through js anyways so you will always have js code in the loop here.
You literally, and I mean literally read my mind dear random stranger down to the wording. Micro is definitely underrated.
Micro is a truly goated software. I mean, it can genuinely replace vscode for small scale editing in the context of shopify that the parent comment was referring to.
It also helped me in physics when I had to remember the units like 10^-6 being micro, 10^-9 being nano etc. and the funny thing is that I used to remember it in the start by seeing I am not sure if it was on micro's github or something but it was a comment on how micro has more features than nano and thus it's name.
So like for some time I definitely felt like I was thinking of micro software, then nano and making the feature comparison to find micro to be larger than nano.
Might seem kinda niche but I ABSOLUTELY LOVE MICRO. Its the one software that I install everywhere, even on my android phone by using UserLand[1] with alpine linux.
I tried writing python code on my phone and it was definitely pleasant thanks to micro.
I'm excited for libghostty. Ghostty has a few design decisions I'm not a fan of, but being able to make my own terminal program or embed a terminal into any app with libghostty is a potential gamechanger.
On the very day ghostty refuses to load my theme because themes now start with an uppercase letter; the same day I'm no longer able to enter `^` (a caret) for some reason. Not to mention the multiple times where the clipboard suddenly and completely stopped functioning, in the last few weeks.
I love ghostty, but if it keeps suddenly failing for no apparent reason I might have to go back to wezterm.
License: This is a perfect example of a project I think should be LGPL licensed and not MIT. Why? Let's suppose he's successful and libghostty become ubiquitous. What happens then? At some point you will be using a closed source version of it on Windows or MacOS or whatever - how do you know it's not logging in the background? The potential for this exists today in every commercial terminal emulator, but why make it easy to take something ubiquitous and backdoor it? On the flip side, what's a good reason for it to be MIT licensed?
Am I missing something? The LGPL only applies to the library itself--you can dynamically link to it from proprietary code. So in this hypothetical scenario, someone could just write a terminal emulator (or IDE, or what have you) that dynamically links to libghostty and put as much telemetry in it as they wanted, couldn't they?
No, you're not but I did! So GPL then. Maybe I just wanted to make up a scenario. I'm not sure why MIT/BSD have become so popular, they have their place but I don't think they have any place in software infrastructure.
>> I truly don’t understand nor follow what you mean? What makes MIT/BAD bad for software infra?
You should have access to the source for your infra as a user of it. MIT/BSD allow people to deploy without offering source, so they can do whatever they want with it and then you get to use it. TiVoisation is more than possible among many other user-hostile options. "But you can always get the source!" is not a valid response - you can't get the source for this thing that's running right now.
Freedom to deny other peoples freedom is no part of "Free Software" but it goes on all the time with "Open Source".
I love how this guy founds + takes public + sells a multibillion company and then just goes right back to hacking. Legend.
Hashimoto is an absolute wizard, but what I find most compelling about him is his absolutely uncanny ability to segment and abstract systems and interfaces in a way for maximum composability and minimal entanglement. He's like the walking embodiment of Rich Hickey's Simple Made Easy philosophy. It's like he designs software systems in such a way that they have no choice but to operate correctly and predictably.
Also I just tried Ghostty for the first time. With iTerm2 and the Zsh/Powerlevel10k theme, there's an extremely brief but perceptible lag from running a command and the render. In ghostty it feels actually instant.
I only wish after 1.3 he could hand over Ghostty to others for maintenance and move to improve other things.
The software world really needs people like him to drive things forward.
Ghostty is his hobby project: why would he do that?
Because maintaining is not as fun as starting? Because other people can maintain a well structured project but not as many people can start something from scratch? Because his skills can be better used elsewhere?
That sounds like the Google philosophy though, where smart people come up with ideas and write the initial implementation (then get promoted) and other less smart people take over, and you end up with a mess that had great potential like Bazel?
My understanding is that this is not a philosophy at all, it is an incidental result of a bad set of misaligned incentives, where to be promoted you need to start something new or drive growth and post launch maintenance/growth is weighted much less
> Because maintaining is not as fun as starting?
The previous commentor is hoping he'll move on because maintaining isn't as fun as starting? Why do you think he can't decide that for himself?
Don't know his motivations. Just gave mine.
Hmm I found the opposite to be true. Maintaining a well structured project is a much harder skill then starting green field.
How is this not driving things forward? It's the best terminal emulator I've ever used.
The person you replied to agreed, but the maintenance could be done by other people while Hashimoto could continue to improve other technologies.
It's Mitchell's hobby project. If he had an interest in improving other technologies, he would. People are allowed to have hobbies.
That sounds like what Drew DeVault does (did?). He made sway, aerc, and others but largely doesn't work on them now AFAIK, passing off work to other people such as emersion.
They have things for contributors, so if somebody wants to jump on it, they can
I just tried it, coming from Kitty, and I didn't notice many differences. What's good that I can try?
I love ghostty. The only thing that is missing is a find feature which is such an odd omissions for an incredible piece of software.
To add... The missing scrollbars feel painful to me. I don't even want them for scrolling, really. I want the visual reference for:
"How much content is in this window"
and
"Where am I in that content"
I still use it daily but it means I have to switch tools for certain things, and reading log files or log output is one of the more common reasons I switch.
I should probably look into trying to get the scrollback info into my statusline, No idea how easy or hard that is - so if someone has done it, feel free to shoot me pointers.
Suggestion to use Klogg (https://github.com/variar/klogg) for reading log files.
I always was on the camp of "tail -f file.log" but since discovering this app, I saw the light.
Or lnav (https://lnav.org/).
Centralized logging, a la Splunk is the gold (with what they charge, probably platinum) standard here. Watching someone who knows what they're doing at a Splunk console for a distributed system they understand is a thing of beauty. It's just a shame they charge so much.
Same here, scroll bars are literally the one missing feature keeping me from making Ghostty my daily driver. But the release notes roadmap mentions they will add them for 1.3. Waiting with anticipation.
Maybe I'm misunderstanding your request but do you not pipe outputs into a pager? (Less, more, batcat, etc) I feel like I'm probably not understanding the context though.
Sometimes you just have a command write output.
I'm still not quite sure what you mean. You can still {|,>} output from a program, so less and grep still help.
Or do you mean when you inevitably forget? Well then yeah you're at the mercy of your terminal but as others mentioned ghostty has a hack to help as well as some other terminals. But this should also help reinforce why you should pipe more often and write to files (or tee). It sucks but not making the same mistakes in the future and learning better habits will help you write better code and use better practices.
But that's the age old problem of "you can't analyze the data you didn't record" and that's a footgun you'll experience in every programming language, every experiment, and across many parts of life. Better to record and throw it away than not record and regret it.
> "you can't analyze the data you didn't record"
It IS recorded. It's right there in scrollback (Literally the default buffer to record). It's easily accessible with most tooling, including nice scrolling, mouse support, find/search, etc...
Except in Ghostty, it's not so accessible. No find, no scrollbars.
I end up having to dump it into another tool, which at least they make pretty easy (ex: https://ghostty.org/docs/config/keybind/reference#write_scro...).
Although deciding when to do that would be easier if I had a better indicator for just how much scrollback content exists. Ex - if it's 3 pages... I'll just scroll through it. If it's 3000 pages... time to dump to file.
So no - by default I use a pager... just about never. Why would I when I have absolutely everything in scrollback by default 99.9% of the time?
---
Don't confuse your preferences with "correct" :P
I'm doing just fine with code and best practices, I'm simply stating that this is a rough edge on an otherwise lovely tool.
In that case are scroll bars even helpful for determining position?
My terminal history is normally huge, and the log output would be some unknown percentage of the total scrollable history.
In the latest version Shift + ctrl + alt + j copies the entire screen and opens it in what I assume is $EDITOR (maybe this can be configured). Or just shift + ctrl + j to copy the screen to temp file and pate the path. Either way, it then just becomes configuring it to use a pager by default, or `shift+ctrl+j` and then | less. This is similar to how kitty does search, by essentially outsourcing it to an external pager. I was put off by it at first, but it's very much in the UNIX philosophy.
Kitty at least let's you control the program used, and even set up different bindings for different programs.
I find the search through a file actually makes sense.
If you don't know, you can send the log to a file, and open that file to look through it. More powerful than just a search next, as you can have instance counts, search with regexp and all the bells and whistles and it virtually stops the logging.
At first I thought the same as you, now I've become quite partial to this concept. I hope they don't remove this.
They'll be working on it soon: https://x.com/mitchellh/status/1964826929326153898
1.3 is slated for release in March 2026.
There's a few things like this but I'll give a pass because how new it is. Were this a mature project I'd be upset at incompetence, but given how young it is I think it's too early to be harsh.
Plus, they're very open about what they're doing and prioritizing. As another commenter said, it's coming soon. For the rest, open feature requests, you might have needs that others didn't think of or even realize they needed
I honestly didn’t even realize it was missing until I was reading this thread, cause I’m always in a zellij session which has a “search in default editor” key.
Pop the whole scrollback in helix, where I can search, select, jump around, paste stuff into a scratch buffer. It’s slick. It’s got a normal search too. But yeah I haven’t used a built in emulator search in a while I guess!
He has the skills and the time to make the best possible software and he's doin it for free Respect for that.
The nerds would do sooo much better for ourselves if we could build a momentum behind comments like this.
Finding mechanistic (& programmatic) sympathy counts for so so much! Shapes the arch of software so much! But it's such invisible unknown work to most people, not so overtly clear & obvious but something that constantly builds day after day, person after person, incremental 0.2% gains compounded by lack of friction.
As well as just promoting good practioners, it feels like discourse about software architecture has really fallen off. We are deep inside rabbit holes specific to this framework or that, and there's only occasional popping out to free air to bring back some observations from the burrows. Ideally we'd have many more volumes of Architecture of Open Source Applications (2011), for example, to really dive into what is, to give us some common referents to learn from and talk about. https://aosabook.org/en/
This is all so core, so worth getting deep on & looking how things are assembled, what the interfaces and modules and shapes look like, what the tradeoffs were. But it remains chiefly an arcane art, one that most developers much less most businesses haven't developed a refinement or taste for.
I just want to say, what a dream. To have wealth and be able to create projects for the sake of going the projects. To not have to make the concessions of quality for profits.
There's an old Knuth quote:
And I think we're seeing more and more that these projects made with love are successful. That without the hyper fixation on money we can build good projects that make big changes in a world.In some sense I'm a bit envious of Mitchell but truthfully these types of things make me more question how we've constructed our society and economy. It shouldn't require one to start with wealth to be able to build things that have such an impact. What needs to be changed where we can live up to what Knuth proclaimed. I'm sure all of us have had experiences where were we given the time (and usually not much) we could make things so much better. But we make many sacrifices when we rush. Which leads to more good advice by Knuth
At what point do we push back? We see that the people we really look up to did things so differently. Knuth himself expressed how detail obsessed he was, and such a claim is common among the grey breads. Of course, things change, but are we creating a world with no wizards? Are we creating a world where we reward people for solving problems and making our lives easier? Or are we just maximizing some score of a pointless game?I'd love to live in a world with a thousand more Mitchells, following their passions without the burden of needing to justify decisions to a board who has no interest in quality. How do we create that world?
Like you said, what a legend. But, how do we make more legends?
What we are missing is a sponsorship model that is able to fun this kind of thing. It's not infeasible - look at university grants dispersement as an example of the overwrought predecessor of this. A ton of money moves through such institutions, and funds a lot of interesting projects.
One thing I've been envisioning is something like a "certified B corporation" style qualification that companies can get that indicates they contribute financially back to open source commensurate to the amount of it they consume to run their core business. If everything you do runs through open source software, in a moral sense, one can make the argument that you owe something back to it.
Having recently come out of grad school, tell me about it...
But the same general problem exists in industry. Our fear of doing things non-optimally only results in a less optimal solution. It's a risky move to take no risks.
In both academia and industry you see the same people doing the same things in the same way. It's no wonder things don't change. You can't have a paradigm shift by following the paradigm (playing it safe). I feel like this is a big shame in both tech and academia as the histories of these have always been made by those who rocked the boat. At some point we just have to admit we're not very good at predicting the future and instead of trying to predict what will be the most successful we should fund passion. I'm sure charlatans will get funded too, but its not like we're doing a good job at preventing that from happening now anyways...
Also, in real life, he's a genuine nice guy. I had lunch with him and Armon at a conference when they worked at Kiip, before they founded Hashicorp. Mitchell (to me) was the proper hacker. He just loved everything about computing. This particular conference was about distributed systems and he was just geeking out on everything.
I walked away thinking that no matter what they did, they'd probably be successful. I was extremely happy to find Ghostty and have been using it ever since.
And hacking on tty software too, the geekiest corner of the Unix tech stack.
I love ghostty and use it daily, somehow I missed it was by Mitchell Hashimoto! Very cool to see.
I had tried it briefly previously but in the last couple months I think I have made the permanent switch from iterm2. It's so much snappier and simpler and also reliably handles text reflowing when a long line wraps, which was a constant problem I had in iterm2, where it would insert fake newlines when copying out text that was wrapped.
I also like that I can have my config in a little plaintext file and just drop it onto a new computer and get the same keybindings. I am using the terminator keybindings for creating and navigating between split panes.
Basically MariaDB. He (original creator of MySQL) sold MySQL for a Billion dollars (iirc) forked it, created MariaDB and kept at it. It's been what 15 years now?
Wrong person. You're thinking of Michael “Monty” Widenius, not Mitchell Hashimoto
I think op is telling a similar story, not necessarily the same person.
You are correct. I tried to rewrite my statement after I was being downvoted. English is not my first language to the surprise of most.
100% Legend love Ghostty btw :)
You made me look into HashiCorp again. It's a great company from a developers perspective, in terms of the many useful tools created: Vagrant, Packer, Nomad, Consul, etc
However their financials are... LOL
Revenue US$583 million (2024)
Operating income US$−254 million (2024)
50% loss margin :-)))
This can't be right. How are they still afloat?
Actually, that's the question (or rather the feeling) I had even before seeing these numbers. Just by reading the docs and looking at everything they built (and how they built it) made me wonder if they spend more than they make. That's a really funny feeling I never had before. Like, surely economies can't be that forgiving when it comes to polishing things. Is that what's happening to the company? Mere overspending?
They’re owned by IBM
More like IBM saved them.
Wow, I wonder what his everyday life must be like.
Yeah, it is so hard to be born and raised in California.
I cannot imagine what he went through.
Hero.
Related: https://news.ycombinator.com/item?id=45056901
From the author himself: "My company had an exit, I did well financially. This is not a secret. I'm extremely privileged and thankful for it.".
Honestly, good for him [for doing well financially] and for admitting that he is privileged, and above all, being thankful for it.
This is super cool, it will be so nice to have a truly omni-platform terminal emulator that can even conceivably extend to iOS and Android.
Aside: I didn't realize Ghostty was written in Zig, wow. The first Zig-thing I'm aware of using on a regular basis. It's amusing the repository structure looks exactly like a Golang layout, haha.
https://github.com/ghostty-org/ghostty
It doesn’t (look like a go layout) and that’s a good thing because pkg/ src/ etc aren’t good go directories.
Since go 1.25 you can declare the what is the “root” of a go module, which is a good thing. Not everyone wants the top level directory filled with go files specially in a multi language repo. So this idea that /src is not good go directory will soon be outdated.
The passion and attention to detail Mitchell has for developer experience is beyond me. I remember the first time I used Vagrant (2011, Santa Monica California) like it was yesterday it was that impactful.
I never thought in a million years I would even think of ditching iTerm2 but when Ghostty dropped I installed it and fell in love.
What exactly does it do better than iTerm2? I gave it a try a few weeks ago and wasn't at all impressed, but maybe I just don't know what I'm missing.
Great question. For me it feels faster (anecdotally) and certain TUI components don’t break. It also is probably my bias of loving his work as well as everyone at work using it. Hive mentality I suppose.
My experience is similar. Stuff doesn't break and it feels extremely snappy. I've switched to Zed for the same reason. I came to realize that I work much better with faster, simpler software, even if it's missing some nice-to-haves. It feels like you're closer to the problem you're trying to solve.
I really want to like Ghostty but:
- Still no ⌘F for find.
- No way (that I know of) to select previous output or specific string and copy with only keyboard shortcuts.
- No ⌘. sending CTRL-C (muscle memory and being advertised as native to the Mac is what one would expect).
- Fonts still don't render as nicely as in Terminal.app. I've fiddled with `font-thicken-strength` and it's close, but not quite. Probably impossible or very hard to replicate due to its Metal rendering nature, but when all you do is look at text all day, it matters.
> No ⌘. sending CTRL-C
This comment sounded familiar[0]. :) For what it's worth, still possible with:
[0]: https://news.ycombinator.com/item?id=42889411Haha, thanks again :)
I've been delaying a major migration to NeoVim on [modern-terminal-I-can-never-decide-which] for years. Wezterm, Ghostty, iTerm2. None is exactly perfect, so I just keep watching them develop.
I recently migrated from vim to neovim and you can just migrate everything. I forgot where I found this but put this in ~/.config/nvim/init.vim
set runtimepath^=~/.vim runtimepath+=~/.vim/after let &packpath = &runtimepath source ~/.vimrcThen in my zshrc (well... I organize differently) I have the function
FWIW, zsh has a commands hash to make stuff like this potentially easier and cleaner. The following isn't quite how I'd do it, but is functionally equivalent.
I wasn't aware of +commands, thanks. Though I'm not sure this is easier and you can translate mine to bash trivially
Yours isn't technically equivalent though it is functionally. If we have both vim and nvim then we'll alias vi twice.
Also, your program provides less clarity. It saves lines but at a large cost to readability. I tend to share my dotfiles with newbies a lot so readability is very meaningful.
I wrap mine up in functions too because at the end of my alias file I can add something like this
It's a bit overkill and I never expect a simple alias like that to fail but there are three distinct advantages I get for being just a tad more verbose:1) I have more complicated versions to deal with things like fd and batcat which have different names different operating systems (`fd` vs `fd-find` / `bat` vs `batcat`) and some additional configurations.
2) I can disable the alias by commenting out one line
3) knowing exactly where the alias is being loaded and thus what aliases are loaded.
Bonus) fails loudly but continues (it's an alias, I don't want you fail fail)
A few extra keystrokes are worth this advantage imo. Especially since I'm using vim so it's actually not any additional typing lol
It's style, so the choice is up to you and they'll accomplish the same things, but I'm just explaining why I use this design pattern. I've only given a stripped down snippet of code so I hope this context helps explain the larger pictures and how it can be used for larger needs.
(Surely this issue must've been discussed/debated elsewhere ad nauseum because it seems an odd design decision to leave out such a common macos binding...)
But having only used ghostty as-is and getting bamboozled by the copy paste situation, this is game changing. I was just going to wait till preferences had a GUI/TUI.. So thanks!
> - No ⌘. sending CTRL-C (muscle memory and being advertised as native to the Mac is what one would expect).
For what it's worth, that shortcut also doesn't work in the Jetbrains terminal, Alacritty, Kitty, and even iTerm2. The only terminal emulator I've tried so far where that works is the built-in Terminal.app.
Search in scroll back is coming in 1.3, unfortunately that's going to be 6 months away.
There's a lot happening in the Ghostty app though, check out the 1.2 release notes
Is there a blog post anywhere that explains why it has taken so long to get search working? Ghostty is such a nice app but that's such a fundamental feature that there has to be a good reason... It's the only thing keeping me on iTerm2.
Try Cmd-Shift-J. It dumps your buffer to a temp file, with the file path made ready for you. I do that and then open the temp file with vi.
That's the workaround I've used. I still want to be able to just press command F and see "error" highlighted throughout the scrollback.
Does nobody use grep?
How can you use grep to search the scrollback/screen buffer?
I don't want to presume your use case, but Ghostty has a command for dumping the buffer to a file, which I use for processing output "too late" to use grep.
keybind = ctrl+alt+shift+o=write_scrollback_file:open
Try a pager instead. Batcat is more feature rich but there's always the good old less (and more) command. Both work great with grep. I do things like the following multiple times a day
Side note / pro tip: on a new line in the terminal press control-x control-e. If you're in zsh you need to edit your config but this will work out of the box for bash.You can leave the cat at home.
https://mrmr.io/cat
but he will feel lonely if he doesn't participate :(
Ghostty has a write_scrollback_file action you can use to write the scrollback to a temporary file. I have this set to cmd+shift+f, so I simply
less [cmd+shift+f] [enter]
You can use grep similarly.
What good is grep if you want to search the build log on your screen that took an hour in the making? Should have thought of tee-ing it, well duh.
You just rawdog that to stdout without a log file?
I do. Setting up log files for each and every command seems tedious. I would rather just cmd+f to search and have it work for everything, as it is a property of the terminal, rather than a specific command.
Of course. Why not? I can use the terminal's search facility to search its infinite scrollback buffer.
If I don't anticipate problems, or things are slower than I thought? Sure
"Do you not have phones?"
Best news I've heard all day.
By the rollercoaster of points this post is experiencing, I may have expressed myself poorly. To be clear, the concept of a modern, cross-platform Terminal core with native GUI implementations for each major vendor is amazing and I'm very glad it exists. It's just not quite there yet for me and I don't know if it ever will be. But I'm very excited by it and hopeful.
It’s weird to me that they consider GTK a “native” implementation — GTK is a whole [opinionated] layer of abstraction on top of the native interfaces.
It’s native in the sense that it’s not Electron, or Java, but it’s still a layer of abstraction which introduces all sorts of quirks.
I was also downvoted for mentioning a few other minor things it is missing as well. While I'm holding out, it's getting there however.
Same here. Mentioned 3 different bugs that happened to me recently, but it looks like it's deemed uncalled for or something.
Interesting, I really prefer the font rendering in Ghostty! I'm with you on the first two items, though.
Despite these shortcomings I still use it daily since the first time I tried it, and have come to really love the project in general. I often won't hop to new software if it offers friction like this, but Ghostty pulled me in despite some snags.
Zed is also wonderful software, though it took me way longer to commit to using it daily.
Should be soon: https://ghostty.org/docs/install/release-notes/1-2-0#roadmap
6 months. 1.3 does not release until March 2026.
Maybe sooner in the nightly builds then :)
A few other ugly quirks in Linux:
- It doesn’t respect the user’s configured fonts. - Clipboard support is inconsistent. - The GTK4 UI is quirky, and has silly slow animations in situations which don’t make sense.
The first two sound like more serious issues, and I’m surprised that others haven’t found them dealbreakers.
Interesting, are there issues on the tracker pertinent to those problems?
I want to like it too. I even joined the Discord I was so excited about it. Unfortunately even at version 1.2.0, which I understand is a total rewrite on Linux, it's only usable for minutes before this bug kicks in https://github.com/ghostty-org/ghostty/issues/2210
> No ⌘. sending CTRL-C
Wait wait wait, this is a shortcut in Mac??
Seriously. I've been using Macs since... ahem... 1985... and did not know this.
It means cancel, stop. You can use it to stop progress bars, dismiss dialog boxes, terminate some well-behaved spinning beach balls, cancel a drag before dropping, etc. It's been on the platform since the classic Mac OS days.
I was wondering why the shortcut for stopping a running program in Xcode was cmd+. Now I know
the second bullet single-handedly keeps me in WezTerm (copy/select mode)
Not to mention that every time you ssh to another server your keyboard stops working correctly.
The solution is on their website:
> infocmp -x xterm-ghostty | ssh YOUR-SERVER -- tic -x -
I mean... I could learn what it does, but no way on earth I am going to be typing something like that every time I want to ssh somewhere.
(This is somehow fixed in 1.2 but it requires additional entry in the config for whatever reason. Can't it just work?)
> (This is somehow fixed in 1.2 but it requires additional entry in the config for whatever reason. Can't it just work?)
"For whatever reason": because it requires modifying `ssh` and doing that by default is a really sketchy thing to do because it is a very security sensitive tool. No program (Ghostty included) should be mucking with that by default. We want users to be aware.
"Can't it just work?": No, because the entire mechanism is flawed, and I didn't invent the mechanism. It's like asking, "why must I drive to my destination, can't I just teleport?" The entire premise of the question is silly, and it's not the car's fault (Ghostty is the car in this example). For those who want to learn more about "the mechanism": read Ghostty's terminfo page, but also just do some light web searching on how terminfo works. Its a total nightmare.
I really wish that weren't the case, I really do. I promise its just as annoying to me as a user and more annoying to me as a maintainer to have to hear people complain about this repeatedly when I'm not doing anything wrong, personally. I'm playing by the rules. The rules are just bullshit.
We'll continue to make enhancements to improve this while we wait for Ghostty's terminfo to propagate to every machine in the world. It will, it'll just take... a long time. Next up we plan on introducing a `ghostty +ssh` command that you can drop-in replace most `ssh` usage with and it'll automagically just work.
Ouch, I did not expect my complaining would get a response from the author!
Just to be clear, I do think Ghostty is amazing piece of software and it's so fast that it's hard to believe, so thank you for the hard work.
> read Ghostty's terminfo page, but also just do some light web searching on how terminfo works. Its a total nightmare.
Oh, yea, I totally get that. I actually _did_ try to understand what the issue is, but gave up on that time-sink rabbit hole and decided I might give it another try once the xterm-ghostty is more popular.
In theory one can set TERMINFO to the actual terminfo string, but sadly openssh (and I suspect other ssh clients/servers) don't pass that on.
Has the ghostty terminfo been added to ncurses?
I use ghostty daily now. Switched recently. In macOS I can re-map now caps lock to be cmd and it works for cmd+c. Everything else is great, sensible defaults and what's not easily customizable.
Gruvbox light theme looks great too.
The fact it's written in Zig is awesome too, if you ever question if Zig is ready, ghostty is your answer to that.
Not seeing myself going back. It's great experience.
Tip: if you combine your ghostty flow with aerospace, it's nearly perfect setup for your keyboard only experience on mac.
Can you tell my why I should consider ghostty? (as a daily Terminal.app user)
It's fast. Like, faster than your brain expects it to be.
I live in vim and after about 30 seconds of checking ghostty out I switched from iTerm2 to ghostty for good. No regerts.
I use terminology today, which is also "fast." The major issue I run into with ghostty vs terminology is that ssh in ghostty is broken due to the terminfo thing. I don't fully understand why ghostty can't pretend to be a different terminal that remotes know about (for now), but I assume there's some reason.
Have you tried the suggestions in https://ghostty.org/docs/help/terminfo#ssh? I don't know what issue you may be experiencing but this solved my issue with using htop in an ssh session.
For another approach, personally I found just doing `TERM=xterm-256color ssh <server>` more practical than the solutions at https://ghostty.org/docs/help/terminfo#ssh
That is essentially the 2nd solution recommended. I think ghostty could do this itself as a hack. If the target binary is ssh (heuristic, but, ok), inject TERM=xterm-256color into env of the child process at launch.
I guess it isn't a huge deal to have every user to modify their ssh_config instead, but it's an ergonomic pain point for many new users.
Copying terminfo around is more inconvenient than I want to put up with for now. Overriding TERM is fine, I guess.
Mostly terminology is just a fairly good terminal, so I have little incentive to switch. :)
Fast. Very fast. Good configuration and theming options too. Literally no lag, even with thousands of lines scrollback. It doesn't get in your way on your day to day terminal activities. Feels natural.
Everyone telling you that it's fast because iTerm2 is slow. Terminal.app is already faster than iTerm2 and on par o very close to most other alternatives in terms of speed.
I also used Terminal until recently and don't use any of the advanced features alternatives provide. The main reason to switch from Terminal.app is truecolor support. The terminfo thing is annoying but I just setenv TERM in ssh config. Better split panes is nice. Configuration in a text file is a matter of taste, but documentation is good.
macOS Tahoe adds true color support to Terminal.app https://www.macrumors.com/2025/06/16/apples-terminal-app-mac...
Other huge projects are Clickhouse and Bun.
I think you’re thinking of TigerBeetle, not Clickhouse, which is a quite performant db with a fascinating simulation-tested story and proof of performance / safety.
ClickHouse is written in C++ (un)fortunately. Although I do think that Zig would fit it quite well too
Yeah ClickHouse is written in C++, though rust is also possible https://clickhouse.com/blog/rust - would be cool to see zig!
Several years ago I was crazy frustrated with how slow 'less' is for some things (regex search and handling long lines) and was slowing down my day to day operations work. I very briefly entertained an idea to write my own pager to fix those problems but it would have been way too complex to do in my spare time. If libghostty existed then it might have been a different story. If libghostty is done well I think it could be an enabler for many types of terminal applications. I'm excited to see where it goes.
I built a macOS app [1] (to fuzzy search Apple Notes) using Ghostty as a base so that I could use its fast rendering for large lists.
I think there is space for an Electron-like framework but for standalone terminal apps. Looking forward to using Ghostty as a library.
[1]: https://enzom.dev/hot-notes/
I've been meaning to build something with this[1], maybe you'll use it before I get to it.
[1]: https://github.com/charmbracelet/bubbletea?tab=readme-ov-fil...
I've been keeping an eye on this in the hopes that text-reflow (edit: including for scrollback) could be solved in Neovim-based terminals [1]. I'm loving the innovation Ghostty is bringing to the terminal space.
1. https://github.com/neovim/neovim/issues/33155
Do you consider yourself a neovim terminal power user?
I tried a while back to invert my workflow (from tmux driving neovim to neovim driving terminals) because I thought it might be easier to only ever have one buffer open for a given file, instead of attempting to open a file in a given pane only to realize that it's already open in a different neovim instance in a different pane.
When I was testing that stuff out I don't think I noticed particular issues with text reflow that would benefit from being solved by swapping to libghostty, rather my pain points were just about how to adjust to the different paradigm. I'd be curious to hear more about someone who is all in on Neovim embedded terminals (and possibly how libghostty might make it better).
I'm all in on Neovim terminals, having a remote development setup means it keeps my terminal with my neovim window (I use nvim-qt).
Also not sure how ghostty would help, haven't noticed text reflowing issues.
It's not bad, a little awkward getting used to:
- you might want a plugin to give you a "persistent" terminal across all tabs
- I still haven't found a way to clear scroll back while a command is running
- I had to set up mappings for easier exiting terminal mode (c-\ c-n really sucks)
- I had to set up events so whenever a terminal buffer is focused it immediately enters insert mode. While I love vim, I've never wanted modal editing in a terminal
I do indeed live in the terminal (all day due to work), but tmux adds too much value for me to do all terminal management in Neovim (tmux session-management being what I use most). I've just encountered too many visual "glitches" in the Neovim terminal to rely on it for everything. That's not to say, however, that I never use the built-in Neovim :terminal.
> I thought it might be easier to only ever have one buffer open for a given file, instead of attempting to open a file in a given pane only to realize that it's already open in a different neovim instance
This is not a problem in my config:
Since `'autoread'` is by default `on` in Neovim, this seamlessly reloads the buffer if the underlying file has been updated on disk.I'd be curious to hear more about how tmux helps you — I tried it and besides keeping a permanent session open on a remote server to me I didn't find much use for it compared to regular terminal tabs
I use it daily locally, and find it amusing how many only think of it as being useful on remote servers (not to invalidate your use-case -- I'm just contrasting my own use). As a precursor, I view UNIX as my IDE, of which tmux is a part: this IDE runs on Windows (WSL2), macOS, Linux, and Android (Termux). That aside, here are a few reasons I find tmux to be useful in this concoction of tools:
- Session management. I've written custom scripts for myself around this (zoxide + fzf). If you want to see how this can be used, look at ThePrimagen's workflow. I don't use his scripts but he has a good demo of how he harnesses sessions.
- Unified scrollback management - easily search the scrollback, yank it, etc. My favorite thing to do is to yank part of the scrollback, then `Prefix+B,=` to list everything I've yanked (think of this like a "clipboard manager" specific to tmux), select an entry, and press `e` to edit it in `$EDITOR`.
- This one might be a stretch, but I tend to try and use only terminal tools (without being utterly insane) because then tmux can be my "tiling window manager" no matter what OS I'm on. Oh, I have to use Windows for work? Not to worry, tmux runs in WSL2, as do most of my preferred tools, so I feel mostly at home even though I normally really dislike Windows.
- It's scriptable. Read `man tmux` and use your imagination!
Notwithstanding any of that, there are cons, the most apparent one being that I am limited to text-based tools this way. An example of this: getting images to work in tmux, though many modern terminal emulators support them, is a huge pane, so I haven't bothered.
For the curious (like me), here’s more info on bullet 1 (session management):
https://www.reddit.com/r/tmux/comments/1ch9tqp/primeagen_tmu...
That's very helpful – thanks for taking the time.
I think my problem is when I realize that I had unsaved changes open in a different neovim instance. If the file was not dirty in any other open neovim instances then I don't have the same problem.
I use ghostty on (my?) air mac that my brother gifted to me which was his previous device for coding / general purpose etc.
Ghostty is really good and I genuinely appreciate this.
Honestly, Might seem like a small thing but the fact that this doesn't even depend on libc seems to be a really great thing to me and I kinda cherish it!
Thanks! Have a nice day!
Why is not depending on a libc a good thing? Aren't they then rolling their own syscall wrappers?
i desperately want to use ghostty but cmd+f support is just such a dealbreaker. excited about this development though!
It's on the roadmap: https://ghostty.org/docs/install/release-notes/1-2-0#roadmap
oh, incredible! i'll keep my eyes peeled and will be switching to it from iTerm the moment it's available
I have heard this from a lot of people, yet here I am, using a terminal that supports this, and have yet to ever do it. Can you help me understand what workflows you depend on Ctrl+f for? I wonder if I am missing something big.
Don't you ever need to search through a program's output; e.g., to find what failed? Otherwise you'll have to remember to tee everything to a file every time you run a command.
You know, I can't remember ever doing this in a terminal. Not because it doesn't sound useful (it sounds VERY useful), it just sort-of never occurred to me as a thing you could do. It's just muscle memory at this point to rerun the command with a pipe to a pager and search there.
I get ya, but this would be more of my case if my terminal didn't have scrollback search available (or if I really wanted to scrutinize the output carefully).
In foot I just ctrl+shift+r and search back the references one-by-one in a mode, which I guess is 90% of my use case for scrollback.
I guess that's what I always do. or grep or awk or sed.. Maybe if a program takes hours to return it's helpful to be able to search the terminal, but otherwise i'm just pushing up with my arrow and `| grep -v UGLY | grep BEAUTIFUL`
|grep or |less
Not after it's run you don't.
You just run it again.
No, “just run it again” doesn’t work.
I can’t always do this.
It could take a long time to run, and my time is not free. The native MacOS terminal has a buffer search, why doesn't ghostty? It is a rudimentary feature.
forever i've been doing the grug brain way of copying the output to notepad++/sublime-text
You can map something like cmd+shift+f to open the entire buffer in your default text editor, this has been sufficient for a lot of folks - myself included - while we wait for native scrollback search to land.
wow yankcrime describing a literal vi yank crime. i'll give it a try this week, that's helpful
I started using it months ago and don't miss that so much, if I really can't rerun the command and pipe it to less/grep. I'll just use the mouse to copy it to the clipboard and then run `xclip -o` (linux) or `pbpaste` (mac) and pipe it to grep. I know there's a keybinding for putting the buffer in a file, but copying with the mouse is faster than finding it.
Also, not sure if this is by default or it picked it up from my old iTerm2 configuration, but cmd+shift+up/down navigates through prompt lines so it's easy to find the start of a long command. My PS1 in zsh is:
Can someone explain why this was not one of the first features added? Who doesn't want to search their history? There must be some complication.
Mitchell raised the issue himself two years ago: https://github.com/ghostty-org/ghostty/issues/189
As someone who lives in the terminal, I can't say I've ever had the need to do that. It's only by reading the comments that I've realised that there's no search in Ghostty.
probably because you never volunteered to code it up.
but really not all features can make it in 1.0
Guessing its "hard to do it right".
You could run tmux inside it and bind cmd f to copy-mode + /
> i desperately want to use ghostty
Is there any particular reason?
iterm2 is slow, uses a ton of memory, and keeps adding weird bloat/AI garbo
Curiosity?
> All of this is packaged up into a single zero-dependency C API (it doesn't even rely on libc), allowing it to be easily embedded into any popular language ecosystem.
@mitchellh That's super exciting. I'm wondering how easy it would be to integrate in the debugger frontend I'm working on. Do you already know what the C API will look like?
I want to like Ghosty, but not being able to search has been a deal breaker. Guess the alternatives to search are use tmux or grep the ghosty log, but tmux and I have never clicked, and searching the log is not user friendly for how often I find myself wanting to search terminal history. Ghosty has a lot of features I would want, but somehow it is missing search. https://www.reddit.com/r/Ghostty/comments/1i0g2ge/how_to_fin...
Maybe this concern doesn't apply to libghostty, as I think search would be more of a UI concern.
I really want a little adapter that can plug into a monitor to make it a dumb terminal. Instantly on, no full OS, just a VT100 or something.
I've looked into it with a PiZero and some HATs but I'd like something made by smarter people. This would be perfect for that.
Ideally just some dip-switchs to set the terminal to emulate and set the display resolution.
Then what would you do? serial port? ssh? what is your terminal connected to if you don’t have an OS?
Serial port or serial port over "ethernet"[1] (technically it's 8p8c[2]).
My router doesn't have a video port, but it does have a dummy terminal port. I had to scrounge a video card for my server to set it up, but it does have a serial port [3]. So that would have been nice.
Also would be nice for a modern remake of dumb terminal with abandoned monitors.
[1] https://www.cabling-design.com/references/pinouts/EIA-TIA-56...
[2] https://en.wikipedia.org/wiki/Modular_connector#8P8C
[3] https://serverfault.com/a/529159
I'm really digging Ghosty's shaders. It finally feels like a gimmicky terminal done right. Want a Fallout or CRT theme without compromising on other features? You can finally have it. It brings back memories of the early days of Linux when everything was fun and not trying to mimic the most bland OS out there. Think Enlightenment before everything got gnomified.
I want to use this to modernize vterm in Emacs. If I could only synchronize the terminal cursor and the Emacs point, and preserve lines as lines, not split them...
I wasn't aware that vterm need modernizing. Happy user of vterm at work and ghostty at home!
vterm flickers a lot with any busier TUI. Claude Code sucks in vterm. I'd also really look forward to using libghostty to try an alternative to vterm in emacs.
I am not sure i understand your problem 100% but Emacs’ tty package can solve a lot of cursor issues in a terminal emulator.
I can’t figure out how to get the visidata frequency analysis histograms to render properly. Some render as blocks as expected but others are those diamond-question marks. I know it’s a silly hang up but it’s what’s keeping me on iterm. It’s seemingly impossible to search for (or at least I haven’t figured out the right keywords) to troubleshoot too.
Sounds like perhaps your main and fallback fonts are missing the Code Points in question.
I am not a ghostty user, but I think it has a setting for that, if you can figure out which font iTerm2 is rendering those glyphs with.
> excellent feature compatibility such as parsing Kitty Graphics Protocol or Tmux Control Mode, and more
Oh oh oh!!!
> https://github.com/ghostty-org/ghostty/issues/1935
awww...
(while ghostty can parse Tmux Control Mode stuff, it doesn't actually implement the full feature. My dependence on iTerm2 remains...)
Ghostty has the capability to parse Tmux control mode (and this blog post is about libghostty-vt). Ghostty the GUI is missing the capability to map that to GUI elements. But Ghostty understands Tmux control mode today: https://github.com/ghostty-org/ghostty/blob/main/src/termina...
I love the idea, and I'm doing something similar in Ruby (which, yes, means it will be far less universal than a compiled library; that's an intentional tradeoff - I don't need it to be universal, and even less so when someone is willing to invest far more effort into something that is universal) inspired by AmigaOS, where a terminal window was easily accessible to any application whether it was started from the terminal or not, by opening a file handle to "CON:" (which meant e.g. scripts could also pipe this directly into a newly opened terminal window).
It's one of a few "lost capabilities" of AmigaOS I'm glad to see getting addressed. You can of course fix this today by using one of a variety of terminal widgets, but the ease of being able to just spin up a full terminal window from "anywhere" was very powerful for simple utilities etc.
Really looking forward to this, writing my own personal editor in Ruby and not having a library for parsing terminal input is leading me to write my own kitty keyboard protocol parser (which ghostty implements) and having to hand-code lookup tables is driving me batty. I really don't like the idea of using an existing TUI framework as none of them actually implement the kitty protocol.
* https://sw.kovidgoyal.net/kitty/keyboard-protocol/#
Have you released the project? I am somewhat curious!
Also I just went from a somewhat (niche?) languages rabbit hole and I refound crystal / crystal is in my mind currently..
So what are your thoughts on using something like crystal instead of ruby because they are similar in what I know and crystal can create a static binary while being (faster?) than ruby
I am just somewhat curious to know!
It's nowhere near ready but I made a gist:
https://gist.github.com/VinceGuidry/b44f02117d452bde7471dae2...
I like crystal but without pure OO and the fact that the community is so far behind Go or even Scala it's hard for me to see a good use for it.
You can create static binaries with ruby, just use mruby. I had a bunch of small scripts in crystal that I replaced with mruby and never noticed the difference. You don't get runtime reflection or ruby's enormous library base, but it could fill the need. I may end up transcribing my editor, or perhaps just the parser, to mruby once it's ready.
https://mruby.org/docs/articles/executing-ruby-code-with-mru...
For a fully-static binary, the option you want is at the bottom,
If mruby can make static ruby then that is actually kinda great, I didn't knew that.
I was thinking more of in terms of performance for some similar developer experience
Pardon me but isn't crystal OOP? i feel like it was because I was introduced to crystal lang via some youtuber some long time ago and the comments said that they didn't use because it was OOP
Also if mruby is so great (which it sounds here), then surely most ruby contributers can migrate over to mruby if that has these gains. I think mruby compiles to C as the intermediate right? So there are definitely some limitations as compared to being a interpreted langauge but that can be somewhat mitigated, no?
I am not a ruby expert or a crystal expert. Because If I was, I would be telling you your future (a slight pun ofc of being a future teller but yes I don't know much about it except the puts syntax or that 100 seconds video from fireship which gives me 10 year experience in ruby :p )
Crystal is object-oriented, but isn't a pure object system. In a pure object system, literally everything is, or can be turned into an object. Smalltalk is the only other example of a language family that offers a pure object system.
With a pure object system everything is accessible by the system at runtime. It's ruby's biggest strength, but also the superpowers it grants aren't always used for good, which is why a lot of coders hate working in it. I only code for myself, my coding career is behind me, so I don't have to worry about other people's code.
Mruby's big downside is precisely that it's compiled, so the library base available is very small, and there's not much point in building on it, though a few projects have, such as DragonRuby. It also doesn't have a pure object system. Ruby has thousands of gems, mruby dozens, and you can see them all on this page: https://mruby.org/libraries/
https://dragonruby.org/
Static typing is also available in ruby but I understand it's clunky.
It is Ghostty that makes me believe Zig is ready for industry, and I also want to make a game engine targeting Android and iOS first, however, JNI is really annoying and I hate Google Android. I have tried Bevy for Android and as you can see it's not the fault of game engine but Google. So, use Unity or challenge it. Just do it, thanks Gostty!
How easy would it be to implement a tmux shim application that just has line numbers natively in the pty itself for copying?
Tmux copy mode is already great, my one gripe is no line numbers.
This script https://gist.github.com/Nimmidev/2cf4d5cc80dce32d0240ec7b3cf... is pretty good, but I still get frequent bugs with it, and also it just doesn't work in fullscreen mode (2 panes).
The core issue is that it's allocating a new tmux pane with the sole goal of mirroring line numbers; it would be nice if they synced up in the same pane, avoiding the above issue.
Piping it into neovim is an option that you can do on both neovim and zellij. zellij loses colors, and neovim is probably the best solution to this problem but then again I don't want to have to remember to turn on/off line numbers every time and I personally like one-off panes. Separation of responsibilities, I guess.
Long-winded rant to basically say: would a standard like this solve my issue easier? From what I understand of terminals, I would need to parse the underlying pty, maintain a scrollback buffer internally in the wrapper shim, and also be able to dynamically adjust toggling line numbers on/off.
If I'm doing this kind of translation, how "leaky" will the abstraction be until I'm basically re-implementing the logic in my middle layer, assuming that "for free" I can get the translation both in and out from the pty?
I've been trying to look closer at TUI tools, but that's what really bothered me. Given just how god awful the VT protocol is, you could get the state machine parsing correct, but the developer still has to learn basically every little quirk that was added over the years, no?
(And before someone makes a false equivalence, no, this isn't the case even with languages like c++ - I'm still learning quirks about it to this day, but I don't have to learn the entire thing to build proper, robust code. It does not seem the same with something like the VT protocol. So yes, I'm aware that some learning should take place, but I'm wondering how structured of a developer experience this will end up being.)
Someone call @keithw. He put a tremendous amount of brainpower into getting the Unicode side of this puzzle "right" (or, as "right" as it is possible to be) when he wrote mosh. I'm sure he and Hashimoto could have a grand old (but zero-width non-joining) conversation!
Through TFA I ended up perusing Hashimoto's blog, and from that moved on to "The Tao of HashiCorp".
Curiously, it hasn't been discussed on HN before, so I started a thread to get HNers' thoughts on it:
https://news.ycombinator.com/item?id=45351378
Ghostty is amazing. The attention to detail means that the basics work right, which I couldn't say about any other "modern" terminal apps. I still wish for xterm-like text selection, but otherwise this is the best terminal out there, at least on macOS (Apple Terminal.app is second).
I would love to use Ghostty but it renders fonts so ugly on external monitors its impossible to use https://github.com/ghostty-org/ghostty/issues/4504
idk why i am the only person that dislikes ghostty, i just have a tendency to put my machines under heavy load and keep using the cli and ghostty chugs super super badly when there's any gpu or cpu load, especially compared to "slow" terminals. zero issues with iterm2 on macos or gnome terminal on linux under load but that plus zero bells or whistles makes ghostty seem like a purposeless toy to me.
Terminal and shell are becoming more and more important with the advent of coding agents, the graphical progress bar is a super nice touch too
How does this compare to https://gitlab.gnome.org/GNOME/vte/ ?
Vte is GTK-only, while libghostty aims to be cross-platform.
The ghostty version probably doesn't write the entire scrollback buffer to disk (given its lack of dependencies).
For that matter: how does it compare to libvterm (https://www.leonerd.org.uk/code/libvterm/)?
libvterm is great. Ghostty supports many more features, but the most important I hear from other libvterm users are that it's missing scrollback and reflow on resize, which are both pretty major pieces of functionality.
Example: Neovim is considering the switch to libghostty-vt when its ready. https://github.com/neovim/neovim/issues/33155
libghostty will eventually export a Terminal widget that can be used in GTK as well, so in that regard they serve a similar purpose.
Will there also be a React component? Or is this not intended for web frontends? I was wondering, because Vercel et al. were mentioned in the beginning.
They mention it's compiled to WASM.
I have a confession: I don't know the difference between a console, terminal (-emulator?!?), shell, bash and all these things. It's all just a black window with text for me. I've never understood why people talk so much about git shell when my normal windows cmd/Powershell can do just the same git commands. I'm also a prolific and successful software engineer. These two things shouldn't mix. So please tell me what I missed in life when these things are completely alien to me.
I think that you might enjoy this (paid) zine: https://store.wizardzines.com/products/terminal
> The terminal seems simple at first (you type in commands and run them! no big deal!), but the more you learn, the more you notice a million little inconsistencies (why does pressing the arrow keys sometimes print out "^[[D"? why does selecting text sometimes not work? why are the colours sometimes unreadable?) that make it feel like an inscrutable black box. And it often doesn’t feel worth it to learn more because documentation about the terminal is so fragmented and full of obscure jargon.
> But! Understanding just a little more about the terminal can make your experience WAY better. You can quickly recognize what’s causing a problem (“oh, my arrow keys aren’t working because this is one of those annoying REPLs that doesn’t support arrow keys!”) and immediately fix it (“I’ll use rlwrap!). Or you can turn “wow, this text is unreadable” into “oh, my terminal emulator is responsible for colours! I’ll just go into the settings and reconfigure my colours!”.
Old but amazing good small series.
https://youtu.be/07Q9oqNLXB4?si=FthNcZSYQSNnT0mP
Kind of surprised that we haven’t created this library before now. But glad to see it coming.
I recently ditched VS Code in favor of Neovim and thanks to Ghostty transition has been a success. As a Mac user I use the cmd key in tons of shortcuts and it just worked out of the box, no need to send weird escape sequences
Total newb here - but lets say i want to make a terminal based video game (think old school text / visuals / menus), will something like Libghostty be something relevant to that?
To expand on the other replies: There are two very different tasks here.
This library would be for the task of taking a received ECMA-48 character stream and turning that into a series of actions on some sort of model of a terminal's display, keyboard, and mouse.
You are looking at a library for the (roughly) inverse task of taking a model of UI widgets and turning that into a transmitted ECMA-48 character stream.
Here's an example of the former:
* https://github.com/jdebp/nosh/blob/trunk/source/ECMA48Decode...
* https://github.com/jdebp/nosh/blob/trunk/source/SoftTerm.h
Here's an example of the latter:
* https://github.com/jdebp/nosh/blob/trunk/source/ECMA48Output...
* https://github.com/jdebp/nosh/blob/trunk/source/TUIDisplayCo...
* https://github.com/jdebp/nosh/blob/trunk/source/TUIVIO.h
* https://github.com/jdebp/nosh/blob/trunk/source/TUIVIOWidget...
No, you would use ncurses, or a library built on it (like Curses Development Kit, CDK)
I imagine if you wanted to make a self-contained application so non-command-line users can play it, then maybe you could use it. Otherwise, probably out of scope of the game itself.
This is for terminal/ttys, this isn't something you'd use for game dev. There are tons of other tools out there for that
ooh a web version would be insanely cool.
Sounds like a feature-complete successor to libtmt:
https://github.com/deadpixi/libtmt
Such incredible news. I'm absolutely going to use this.
How does Ghostty compares to kitty and alacrity?
How is he such an absolute chad of a developer?
He has side projects that would make careers for most of us.
Mitchell and bellard and to somewhat degree fwsgonzo (you might not know him but I absolutely like his work on libriscv, the game he's working and the kvm sandbox which he's working on too) are my definitions of literal coding chads.
People should really measure coding proficiency as being 10% of these guys or some fraction instead of whatever metric they are using now (slightly kidding ofc)
They are the 10x developers and I feel like this sentiment is echoed all across the industry
Ive worked with 10x people before. All these people you mention and Torvalds too seem to be 100x developers.
How could I've forgotten the legendary torvalds!
Sqlite team also seems to be one of the 100x people imo. Maybe I just cherish sqlite's simplicity and its hyperfocus and it delievers exactly what it says (somewhat), a lot!
Definitely feel like we need a forum where we can all compile a list of 100x developers and vote who is the best 100x developer
I feel like although Mitchell is up in the top three. To me bellard/torvalds tops it all of if you consider it from a pure software perspective and not that Mitchell went back from a billion $ empire to hacking again
Bellard has created qemu and ffmpeg and tcc/jslinux/ I think even xterm js kind of came somehow due to bellard so a lot of terminal emulators and their recent work is just as fascinating.
Torvalds with linux and git are two back to back things which we know and love and use today.
Bellard is just a little less known because unlike linus torvalds who named the os as linux ie. after himself. Ffmpeg and qemu don't really tell them that they are created by the same guy.
I thought ‘chad’ was a pejorative; is Chad good again? :)
I though it just means an exceptional alpha male type. I didn't know it was ever pejorative.
Super exciting. I hope to be able to embed a Ghostty-backed terminal into my Rust app in the future. Amazing work as always, Mitchell!
I enjoy ghostty and zellij. Zig and rust built.
I was really excited to try ghostty, but the text looks blurry when I place it on my mac's external monitor
I love the idea of libghostty making terminal tunnels cleaner and less leaky, because I hate messy pipes.
Interesting!
The nix package for ghostty is broken/unavailable on MacOS at the moment, but installing the brew cask (`brew install --cask ghostty`) and a basic config file:
```
font-family = Monaco
theme = dark:Catppuccin Frappe,light:Catppuccin Latte
```
Got me almost everything I used from iTerm2. Nerdfont just worked (in iTerm I think you have to use the `Use a different font for non-ASCII text` setting)
Offtopic, formatting related, and always a fun discussion:
The function signature of [1] ...
(^^^ probably not readable in small screens)... is the perfect example of why "align continuation lines to the open bracket" is the absolute worse option versus the simpler, consistent and only logical choice of "align continuation lines with an extra indentation", aka
in Clang-Format [2].[1]: https://github.com/ghostty-org/ghostty/blob/f97518cc10059918...
[2]: https://clang.llvm.org/docs/ClangFormatStyleOptions.html#ali...
Indeed. Also note this isn't the real public C API. This is, as I noted in the blog post as a disclaimer, an internal-only C API so it is admitedly very ugly.
(I assume you know this, just adding context for other readers)
Yep, sorry in my mind I was going to mention it (with "even if it's only me seeing private code like this" or something similar) and later forgot after some edits to fix the formatting...
Anyway, I hope it doesn't get lost that this comment was only meant to be half informative, half public statement, and half a light joke :)
Great project for anyone creating a hobbyist OS.
I love this. I've been thinking about writing a clone of SecureCRT (great terminal emulator, sadly greatly expensive as well) for years.
If I could "just" use libghostty, that would be great!
I wish they had used something like this in Shopify, instead they decided it was a good idea to embed VScode in the browser in case users want to make a small modification to theme files, it even has a 5 second boot time.
What’s a user-friendly mouse friendly modeless cli text editor you can compile to wasm?
I don't think its really practical to do so. wasm can't directly access the canvas or webgpu contexts. it has to route date through js anyways so you will always have js code in the loop here.
micro?
You literally, and I mean literally read my mind dear random stranger down to the wording. Micro is definitely underrated.
Micro is a truly goated software. I mean, it can genuinely replace vscode for small scale editing in the context of shopify that the parent comment was referring to.
https://micro-editor.github.io/
It also helped me in physics when I had to remember the units like 10^-6 being micro, 10^-9 being nano etc. and the funny thing is that I used to remember it in the start by seeing I am not sure if it was on micro's github or something but it was a comment on how micro has more features than nano and thus it's name.
So like for some time I definitely felt like I was thinking of micro software, then nano and making the feature comparison to find micro to be larger than nano.
Might seem kinda niche but I ABSOLUTELY LOVE MICRO. Its the one software that I install everywhere, even on my android phone by using UserLand[1] with alpine linux.
I tried writing python code on my phone and it was definitely pleasant thanks to micro.
[1]:https://github.com/CypherpunkArmory/UserLAnd
ghostty and atuin are two of the first things I put onto any new macOS or Linux machine I use
Newb question: what’s the difference between a terminal and terminal emulator? I’ve never figured that out
The former is a piece of hardware from the 70s, the latter is software that emulates the former.
I don’t understand as 1970s hardware requires firmware/software to take input/output with the user?
WezTerm seems to be missing from the list, and is a superset of those listed.
I'm excited for libghostty. Ghostty has a few design decisions I'm not a fan of, but being able to make my own terminal program or embed a terminal into any app with libghostty is a potential gamechanger.
problem: we have ten different VT terminal parsers
solution: write a new VT terminal parser to replace the other ten
result: we have eleven different VT terminal parsers
Mandatory reference
https://xkcd.com/927/
On the very day ghostty refuses to load my theme because themes now start with an uppercase letter; the same day I'm no longer able to enter `^` (a caret) for some reason. Not to mention the multiple times where the clipboard suddenly and completely stopped functioning, in the last few weeks.
I love ghostty, but if it keeps suddenly failing for no apparent reason I might have to go back to wezterm.
Chill, there was literally one single update since the initial launch. It's not like it's getting breaking changes every day.
You seem to be using tip. Switch to using a stable release (1.2 is the latest one) instead if you don't want to be exposed to potential issues.
FWIW I've using tip since the closed beta and never had major issues.
damn, you should ask for a refund.
Sorry, I hadn't realized only positive feedback was allowed here.
License: This is a perfect example of a project I think should be LGPL licensed and not MIT. Why? Let's suppose he's successful and libghostty become ubiquitous. What happens then? At some point you will be using a closed source version of it on Windows or MacOS or whatever - how do you know it's not logging in the background? The potential for this exists today in every commercial terminal emulator, but why make it easy to take something ubiquitous and backdoor it? On the flip side, what's a good reason for it to be MIT licensed?
Am I missing something? The LGPL only applies to the library itself--you can dynamically link to it from proprietary code. So in this hypothetical scenario, someone could just write a terminal emulator (or IDE, or what have you) that dynamically links to libghostty and put as much telemetry in it as they wanted, couldn't they?
>> Am I missing something?
No, you're not but I did! So GPL then. Maybe I just wanted to make up a scenario. I'm not sure why MIT/BSD have become so popular, they have their place but I don't think they have any place in software infrastructure.
I truly don’t understand nor follow what you mean? What makes MIT/BAD bad for software infra?
>> I truly don’t understand nor follow what you mean? What makes MIT/BAD bad for software infra?
You should have access to the source for your infra as a user of it. MIT/BSD allow people to deploy without offering source, so they can do whatever they want with it and then you get to use it. TiVoisation is more than possible among many other user-hostile options. "But you can always get the source!" is not a valid response - you can't get the source for this thing that's running right now.
Freedom to deny other peoples freedom is no part of "Free Software" but it goes on all the time with "Open Source".