To promote this to a top-level comment, because I think it's important context: nothing in the README file says the point is to be "beginner-friendly". As far as I can tell, that's something that Hacker News or whoever submitted this used as a description. Neither the word "beginner" nor the word "friendly" appear in the README; its goal is to "be the smallest functional Emacs in less than 2000 lines of C."
I think somebody incorrectly described Atto as comparable to nano/pico, and these comments are dragging it because it's not something that it was never intended as.
I would never drag pico. I wrote software that's been packaged in Linux distros in pico. And I lost the urge to be The Smartest Guy In The [Room/Mailing List/Forum] a long, long time ago.
True, many a junior dev I've had to work with was convinced they're the smartest person in the room, good at what they do but so full of themselves they're unable to adapt, acquire new knowledge or consider new paradigms.
Learning requires some amount of humbleness, to accept you don't know everything.
Not only that, but you constantly feel like Idiocracy is playing out in real time, and that everyone else around you is closer to monkeys than a rational empathetic human... it's exhausting, lonely and debilitating. Then you realize there are people much smarter than you, and you wonder how they must feel.
Reminds me of that House MD episode where the delivery driver was a secret genius who was abusing DXM to dumb himself down just to feel happy.
"For with much wisdom comes much sorrow; the more knowledge, the more grief."
> To ease the transition for new users, #NetBSD -current now includes an alternative editor to vi(1) and xedit(1) in its base system, atto(1). https://man.netbsd.org/atto.1
> It is a modeless screen-based editor similar to nano, pico, xedit, or emacs. It is only 48 kilobytes, making it suitable for inclusion in the rescue system.
I think it's fair to call it beginner-friendly, in context of BSD and vi, but maybe indeed it shouldn't be a part of the title.
I can't imagine a beginner really knowing what is going on in nano.
Every time I get launched into nano/pico on a new system I find it very confusing. To use the help menu you need to know that ^-X means Ctrl-X and M-X means Alt-X. Why "Write Out" and not "Save"? And when I "Write Out" I'm offered DOS format or Mac format, and options to Append or Prepend (not even sure what that does).
It's completely bizarre if you're used to something like Notepad or Word.
> I can't imagine a beginner really knowing what is going on in nano.
When you launch nano without a file loaded, there's a notice at the bottom of the screen: [ Welcome to nano. For basic help, type Ctrl+G. ]
Press 'Ctrl + G' and it takes you to a help page that says, among other things, "Shortcuts are written as follows: Control-key sequences are notated with a '^' and can be entered either by using the Ctrl key or pressing the Esc key twice. Meta-key sequences are notated with 'M-' and can be entered using either the Alt, Cmd, or Esc key, depending on your keyboard setup."
You can't force a user to read docs or learn how the programs they're using work, but this is a reasonable attempt.
Well nano pico atto ... the name makes it comparable. Why would you expect people to not draw that connection?
I think the main take-away is that vi/vim is too complicated for many folks. I use nano a lot on the commandline. For quick editing it is so much more convenient than vim.
I've used mg and find it generally useful. It's my default editor pretty much everywhere.
However, the code is a mess, with deliberately cryptic identifiers (e.g., `w_doto` for `cursorOffset`), redundant functions (e.g., `vtputc()` is a strict subset of the behavior of `vtpute()`) and a bunch of other weird choices (e.g., deeply-baked assumptions that one byte == one char == one column, even thought Big5 and ShiftJIS were already a thing when it was written).
I tried adding UTF-8 support to mg a couple years ago and basically gave up.
I always thought the essential feature of an emacs was its extensibility. I never felt that emacs was just a collection of keybinds and split buffer windows.
Note, I'm not talking about Lisp per se, just some kind of mechanic within the editor to allow the user to make extensions to it.
I think there are a lot of people who are not comfortable with vi. Its much the same reason that some Linux distros default to nano.
I spend a fair bit of time on Linux command lines, but I use a mix of different TUI and GUI editors (including ones that work over ssh for editing files on servers), and none of those are vi. I can manage vi if I must, but its an effort and distracts from the task because I have to think about how to do things in vi.
You can ask Claude to cross-complie, prepare an image (aka install) and run an emulator. I tried exactly this yesterday and in about an hour I could attach to tmux and run tests on emulated riscv64 board.
Oh no! What does that mean for using Wine to play games? From what I understand a lot of Windows games (especially older ones) require 32-bit Wine, which requires 32-bit libs (which I assume requires 32-bit Kernel support). It's the sole reason I run 64- and 32-bit multilib on my Arch box: playing Windows games.
64 bit Linux can run 32 bit programs just like Windows. Modern Wine is using a WoW64 model where the 64 bit host code loads 32 bit DLLs when some app needs them.
Who knew a little emacs-like editor would be so controversial.
I think it looks great. Syntax highlighting, multi-buffer, splits, standard emacs commands... For new users in the emacs camp, this seems like a nice editor to include in the base install. Even as a vi user, I'd be pretty happy being stuck with this in a rescue system.
nano(1) is GPLv3 last time I checked and that imposes more restrictions than most (all?) of the rest of the code in NetBSD. It is not about being "reflexive", it is a violation of the very ethos of the project. Gnu rejecting non-free licenses from their own distributions is not "reflexive", it is simply not in line with what they want to achieve and one should respect that.
Unless they modify it, the GPL really doesn't impact them. AFAIK FSF considers most BSD/MIT style licences to be compatible. It's the other way around where things gets stuck. Unless they need Linux GPU/WIFI/BT drivers to make things work, of course. Then it's fine.
Modification is the whole point of open source: "Unless you want to modify Windows, the closeness does not impact you". If you want to learn about how NetBSD looks at things, there is this page:
As for being compatible. Yes, FSF considers anything permissive to be compatible with their family of licenses. However, it is a one-way relationship in that permissive code can be freely used by GPL projects, but not the other way around as using GPL code would impose further restrictions compared to a permissive license.
To be beginner friendly they should call it text editor. Or notepad, or quickwriter. No beginner is going to guess that something called Atto lets you write words on a computer.
>making it suitable for inclusion in the rescue system
People seem to be missing the point. Also NetBSD has what I call a "real single user mode". Last time I had to boot single user, ed(1) was the only editor available without some work. My guess is this new editor will be useful for people who do not want to deal with ed(1).
I have not seen it yet, but I guess it will be included in 11.1.
The beginner-friendly label seems to be obscuring what atto actually is, which is a very small Emacs-like editor meant to live in the base system. At roughly 48 KB and under 2,000 lines of C, it fits places nano and the larger alternatives do not, especially a rescue environment or single-user mode where you still need something more usable than ed or a stripped-down vi. That constraint also explains the choices people are arguing about here: Emacs bindings come with the design, and a permissively licensed C program can be maintained with the tools NetBSD already ships, unlike a Go editor such as micro. For someone new to NetBSD but already comfortable at a Unix terminal, that is a reasonable trade. It is less a first editor for a general beginner than a safety net that happens to be approachable once you are already in that world.
> Perhaps all that needs to be known to use atto at a basic level is that holding the control key, pressing x, then pressing w will write the currently open file, and holding the control key, pressing x, then pressing c will exit the editor.
Very beginner friendly lol. These people have no idea.
Micro is a beginner-friendly console text editor. This isn't.
Nothing in the README file says the point is to be "beginner-friendly"; as far as I can tell, that's something that Hacker News or whoever submitted this used as a description. Neither the word "beginner" nor the word "friendly" appear in the README; its goal is to "be the smallest fuctional Emacs in less than 2000 lines of C."
tl;dr: I think somebody incorrectly described Atto as comparable to nano/pico, and these comments are dragging it because it's not the thing that it was never meant to be.
Why are they using weird key combinations instead of traditional Ctrl + S, Ctrl + A, Ctrl + C used in desktop applications? Why every console program wants to be a special snowflake and not use standard key combinations, please explain. I do not care that 40 years ago someone used Ctrl + C to send SIGINT, it is 2026 now.
UNIX doesn't exist anymore, and C has lot of negative points (easy to create buffer overflows, integer overflows, use-after-free, and other vulnerabilities).
For many, it is indeed about what they call "muscle memory". Almost impossible to change after so many years. And therefore I hope everyone would provide keybinding alternatives generally (shouldn't be that hard).
These key combinations have been used for 50 years, so I'm not sure why you call them non-standard. They're just a different standard from what you're used to.
To promote this to a top-level comment, because I think it's important context: nothing in the README file says the point is to be "beginner-friendly". As far as I can tell, that's something that Hacker News or whoever submitted this used as a description. Neither the word "beginner" nor the word "friendly" appear in the README; its goal is to "be the smallest functional Emacs in less than 2000 lines of C."
I think somebody incorrectly described Atto as comparable to nano/pico, and these comments are dragging it because it's not something that it was never intended as.
I would never drag pico. I wrote software that's been packaged in Linux distros in pico. And I lost the urge to be The Smartest Guy In The [Room/Mailing List/Forum] a long, long time ago.
what's really the point in being the smartest person in the room? you'll never learn anything!
sometimes humans do irrational things / have illogical behaviour just to glorify their own ego. it happens also to the very best sometimes
Yep, whenever you find yourself being the smartest person in the room, you need to find yourself a smarter room.
Or just learn how to teach.
> Or just learn how to teach.
Most people don't want to learn (in particular not learn sufficiently hard to really improve).
Motivating students to want to learn isn't rocket science, just a rare skill to see.
True, many a junior dev I've had to work with was convinced they're the smartest person in the room, good at what they do but so full of themselves they're unable to adapt, acquire new knowledge or consider new paradigms.
Learning requires some amount of humbleness, to accept you don't know everything.
Not only that, but you constantly feel like Idiocracy is playing out in real time, and that everyone else around you is closer to monkeys than a rational empathetic human... it's exhausting, lonely and debilitating. Then you realize there are people much smarter than you, and you wonder how they must feel.
Reminds me of that House MD episode where the delivery driver was a secret genius who was abusing DXM to dumb himself down just to feel happy.
"For with much wisdom comes much sorrow; the more knowledge, the more grief."
From the announcement
> To ease the transition for new users...
So I would consider it for beginners ("new users") and friendly ("to ease").
OP here, I just paraphrased the announcement:
> To ease the transition for new users, #NetBSD -current now includes an alternative editor to vi(1) and xedit(1) in its base system, atto(1). https://man.netbsd.org/atto.1
> It is a modeless screen-based editor similar to nano, pico, xedit, or emacs. It is only 48 kilobytes, making it suitable for inclusion in the rescue system.
I think it's fair to call it beginner-friendly, in context of BSD and vi, but maybe indeed it shouldn't be a part of the title.
I can't imagine a beginner really knowing what is going on in nano.
Every time I get launched into nano/pico on a new system I find it very confusing. To use the help menu you need to know that ^-X means Ctrl-X and M-X means Alt-X. Why "Write Out" and not "Save"? And when I "Write Out" I'm offered DOS format or Mac format, and options to Append or Prepend (not even sure what that does).
It's completely bizarre if you're used to something like Notepad or Word.
> I can't imagine a beginner really knowing what is going on in nano.
When you launch nano without a file loaded, there's a notice at the bottom of the screen: [ Welcome to nano. For basic help, type Ctrl+G. ]
Press 'Ctrl + G' and it takes you to a help page that says, among other things, "Shortcuts are written as follows: Control-key sequences are notated with a '^' and can be entered either by using the Ctrl key or pressing the Esc key twice. Meta-key sequences are notated with 'M-' and can be entered using either the Alt, Cmd, or Esc key, depending on your keyboard setup."
You can't force a user to read docs or learn how the programs they're using work, but this is a reasonable attempt.
I use nano the same way I use notepad. I open a file, edit it, hit close, and agree to save when it complains to me.
>Why "Write Out" and not "Save"?
At least "Uncut" has been renamed to "Paste".
Ok, we've removed beginner-friendliness from the title above. It's still interesting!
Well nano pico atto ... the name makes it comparable. Why would you expect people to not draw that connection?
I think the main take-away is that vi/vim is too complicated for many folks. I use nano a lot on the commandline. For quick editing it is so much more convenient than vim.
OpenBSD also has an Emacs-y non-extensible editor, mg(1):
https://man.openbsd.org/mg
Just noting for completeness, as I didn’t catch any mention of it.
Also available for most linux distros as a package.
I've used mg and find it generally useful. It's my default editor pretty much everywhere.
However, the code is a mess, with deliberately cryptic identifiers (e.g., `w_doto` for `cursorOffset`), redundant functions (e.g., `vtputc()` is a strict subset of the behavior of `vtpute()`) and a bunch of other weird choices (e.g., deeply-baked assumptions that one byte == one char == one column, even thought Big5 and ShiftJIS were already a thing when it was written).
I tried adding UTF-8 support to mg a couple years ago and basically gave up.
I'll probably be an atto convert.
> https://man.netbsd.org/atto.1
> atto is not the standard editor.
Hah. That is clearly a reference to https://www.gnu.org/fun/jokes/ed-msg.en.html
NetBSD isn't the first to use the name. My pet text editor is also called atto: https://git.levitati.ng/Levitating/atto
And before you was: https://github.com/makuke1234/atto, which notes its name originating from A Truly Tiny text editOr.
Also used by Moodle, the open source LMS: https://docs.moodle.org/400/en/Atto_editor
I always thought the essential feature of an emacs was its extensibility. I never felt that emacs was just a collection of keybinds and split buffer windows.
Note, I'm not talking about Lisp per se, just some kind of mechanic within the editor to allow the user to make extensions to it.
NetBSD has beginners?
Beginners to NetBSD who are not vi users.
That's not an empty set? I'm an eMacs guy, and I use vi when on the shell
I think there are a lot of people who are not comfortable with vi. Its much the same reason that some Linux distros default to nano.
I spend a fair bit of time on Linux command lines, but I use a mix of different TUI and GUI editors (including ones that work over ssh for editing files on servers), and none of those are vi. I can manage vi if I must, but its an effort and distracts from the task because I have to think about how to do things in vi.
You can ask Claude to cross-complie, prepare an image (aka install) and run an emulator. I tried exactly this yesterday and in about an hour I could attach to tmux and run tests on emulated riscv64 board.
Careful, they might ban you!
Currently seeing an influx of interest from new people who want an OS with strong anti-ai stances.
And those who still use 32-bit machines, as Linux is in the process of killing support for them.
Oh no! What does that mean for using Wine to play games? From what I understand a lot of Windows games (especially older ones) require 32-bit Wine, which requires 32-bit libs (which I assume requires 32-bit Kernel support). It's the sole reason I run 64- and 32-bit multilib on my Arch box: playing Windows games.
64 bit Linux can run 32 bit programs just like Windows. Modern Wine is using a WoW64 model where the 64 bit host code loads 32 bit DLLs when some app needs them.
Who knew a little emacs-like editor would be so controversial.
I think it looks great. Syntax highlighting, multi-buffer, splits, standard emacs commands... For new users in the emacs camp, this seems like a nice editor to include in the base install. Even as a vi user, I'd be pretty happy being stuck with this in a rescue system.
The source code and README are here: https://github.com/NetBSD/src/tree/trunk/usr.bin/atto
You are welcome!
The dev calls it "Atto Emacs"
https://github.com/hughbarney/atto
Why not just add nano?
Because the BSD's are reflexively anti GNU.
nano(1) is GPLv3 last time I checked and that imposes more restrictions than most (all?) of the rest of the code in NetBSD. It is not about being "reflexive", it is a violation of the very ethos of the project. Gnu rejecting non-free licenses from their own distributions is not "reflexive", it is simply not in line with what they want to achieve and one should respect that.
Unless they modify it, the GPL really doesn't impact them. AFAIK FSF considers most BSD/MIT style licences to be compatible. It's the other way around where things gets stuck. Unless they need Linux GPU/WIFI/BT drivers to make things work, of course. Then it's fine.
Modification is the whole point of open source: "Unless you want to modify Windows, the closeness does not impact you". If you want to learn about how NetBSD looks at things, there is this page:
https://netbsd.org/about/redistribution.html
As for being compatible. Yes, FSF considers anything permissive to be compatible with their family of licenses. However, it is a one-way relationship in that permissive code can be freely used by GPL projects, but not the other way around as using GPL code would impose further restrictions compared to a permissive license.
> Unless they need Linux GPU/WIFI/BT drivers to make things work, of course. Then it's fine.
The parts of Linux that get integrated into BSDs should not be GPL-licensed, e.g. <https://github.com/torvalds/linux/blob/c339f4010dea8b0d754f8...> and its siblings use MIT.
Size:
Linux:
per the link, atto(1) is 48 kilobytes.
Is this the same editor that is used in the open-source Moodle LMS?
https://docs.moodle.org/400/en/Atto_editor https://github.com/moodlehq/moodle-editor_atto
Moodle's is a PHP-written web editor. Submitted is a C-written terminal editor.
Genuine question: Are there a lot of "beginners" hopping on NetBSD?
I feel like most people who even know about NetBSD would be technical enough to at least function at a basic level in something like vi or nano.
For me the most easy-to-use editor for cli / ssh and relatively lightweight is https://github.com/sinelaw/fresh everybody can just use it intuitively.
I’ll take a look at this. But it’s worth noting that NetBSDs goal was to have something small enough that it can be used in a recovery shell.
I think they'd have your head on a plate if you suggested a vibe coded text editor.
There is nothing intuitive about any text editor.
Badly editorialized title.
To be beginner friendly they should call it text editor. Or notepad, or quickwriter. No beginner is going to guess that something called Atto lets you write words on a computer.
>making it suitable for inclusion in the rescue system
People seem to be missing the point. Also NetBSD has what I call a "real single user mode". Last time I had to boot single user, ed(1) was the only editor available without some work. My guess is this new editor will be useful for people who do not want to deal with ed(1).
I have not seen it yet, but I guess it will be included in 11.1.
I think single user has nvi as well.
Edit: apparently it requires /usr to be mounted though
The beginner-friendly label seems to be obscuring what atto actually is, which is a very small Emacs-like editor meant to live in the base system. At roughly 48 KB and under 2,000 lines of C, it fits places nano and the larger alternatives do not, especially a rescue environment or single-user mode where you still need something more usable than ed or a stripped-down vi. That constraint also explains the choices people are arguing about here: Emacs bindings come with the design, and a permissively licensed C program can be maintained with the tools NetBSD already ships, unlike a Go editor such as micro. For someone new to NetBSD but already comfortable at a Unix terminal, that is a reasonable trade. It is less a first editor for a general beginner than a safety net that happens to be approachable once you are already in that world.
> Perhaps all that needs to be known to use atto at a basic level is that holding the control key, pressing x, then pressing w will write the currently open file, and holding the control key, pressing x, then pressing c will exit the editor.
Very beginner friendly lol. These people have no idea.
Micro is a beginner-friendly console text editor. This isn't.
It's using keystrokes inspired by emacs. Ok.
As we all know, beginners love Emacs keystrokes.
Micro is MIT licensed, so it should be shippable in NetBSD base.
https://github.com/micro-editor/MICRO
But written in Go, and I would expect the NetBSD base only includes software buildable using tools already in the base system.
Good point
11 MB is also a lot bigger than 48k
I do think Micro is a nice editor, but we need a tiny, BSD licensed pico/nano for true beginners
I didn't realize the size difference was so huge.
Atto has a great man page that would give a user everything they needed to know if they were unfamiliar with emacs keybindings.
I don’t think NetBSD and yourself are using the term in the same context.
By this I mean, a beginner to NetBSD is already a far more competent user than a beginner in many other contexts of OS users.
Project headline is “The smallest functional Emacs in less than 2000 lines of C.”
How someone jumped to beginner-friendly from that is the real question.
Nothing in the README file says the point is to be "beginner-friendly"; as far as I can tell, that's something that Hacker News or whoever submitted this used as a description. Neither the word "beginner" nor the word "friendly" appear in the README; its goal is to "be the smallest fuctional Emacs in less than 2000 lines of C."
tl;dr: I think somebody incorrectly described Atto as comparable to nano/pico, and these comments are dragging it because it's not the thing that it was never meant to be.
Why are they using weird key combinations instead of traditional Ctrl + S, Ctrl + A, Ctrl + C used in desktop applications? Why every console program wants to be a special snowflake and not use standard key combinations, please explain. I do not care that 40 years ago someone used Ctrl + C to send SIGINT, it is 2026 now.
That sounds like Emacs. Wouldn't know though, given that I dislike vim too.
I would prefer ms-edit or nano if I have to edit text files from the terminal.
You might enjoy the new Microsoft Edit [1].
[1] https://github.com/microsoft/edit
Atto's default key bindings match Emacs, which is about as standard as you get for non-modal editing keys in older Unix traditions.
and isn't at all beginner friendly
Emacs is probably older than me why should we care about what they used in 80s?
I see you’re using Latin characters...
Depends, do you care about UNIX and C, as products from 1970's?
UNIX doesn't exist anymore, and C has lot of negative points (easy to create buffer overflows, integer overflows, use-after-free, and other vulnerabilities).
It surely does, Linux distros, BSDs, Aix, z/UNIX are all alive and well.
Despite all those flaws, people keep writing new software in C every day, go figure.
It's 2026, why would anyone want to use an Intel architecture?
Or ARM? It's from the 1980s! So old!
I would assert Wordstar style are equally as old.
But are mostly forgotten while emacs keybindings are still popular and found in all kinds of places you wouldn't expect, like MacOS textboxes.
Not when using joe, for example.
> match Emacs
For many, it is indeed about what they call "muscle memory". Almost impossible to change after so many years. And therefore I hope everyone would provide keybinding alternatives generally (shouldn't be that hard).
Because this isn't beginner friendly it's a new micro Emacs and those are standard libreadline/emacs bindings.
These key combinations have been used for 50 years, so I'm not sure why you call them non-standard. They're just a different standard from what you're used to.
I think you're underestimating just how "standard" many of the shortcuts are when mapped against a standard shell.
"Standard shell" also should allow Ctrl + C for copying, it makes no sense to use a different combination for console.
55 years of a standard, but then you come and say that the standard should be something else.
That's… not what "standard" means.
Okay, but now you have to modify the shell and configure a different shortcut for SIGINT without stepping on something else.
Is this problem a Nobel prize level problem?
These are about as standard as you can get when it comes to Unix keybindings
> I do not care that 40 years ago someone used Ctrl + C to send SIGINT, it is 2026 now.
Not "40 years ago". Non-stop all day every day for 55 years.
The fact that you personally don't use it today is as irrelevant as if you were ranting about "why doesn't Linux have a Start menu!? It's 2026!!!"
How is Ctrl+S "more traditional" than C-x C-s? C-x C-s is like 50 years old by now. Surely that qualifies as "traditional".