Hacker News new | past | comments | ask | show | jobs | submit login
Sublime Text 3.0 (sublimetext.com)
1653 points by fbnlsr on Sept 13, 2017 | hide | past | favorite | 675 comments

While I know it might be a little hidden, I'd just like to say I'm really glad Will Bond of Package Control[0] was able to join[1] the Sublime Text team. Having first class support for Package Control in ST is definitely one of the features I value the most out of Sublime Text and IMHO one of the things that still keeps it very competitive with Atom other text editors.

I use packages from ST almost every day and plus I wouldn't have gotten my first job without Will (thanks for hiring me even after seeing all that terrible code I wrote), so I can actually say my life would be a lot different without his influence. I'm really happy he's officially part of the team.

[0] https://packagecontrol.io/

[1] https://www.sublimetext.com/blog/articles/sublime-text-3-bui...

Speaking of, the GitGutter plugin I use just broke on this update.


Doesn't even appear in the installable list now. Hopefully he'll get some of these hickups with package manager sorted.

If it isn't in the installable list, you may have it installed already.

Package Control is open source and contributions are welcome! We finally get to rip out support for ST2 and Python 2.6 now also!

Yeah that was a part of the problem.


It was still installed... but broken.

Weird, it was the first thing I installed after installing 3.0 as a new install. Wonder if it's an update issue in particular.

It's an update issue, I'm talking to a fair few people who had the same issue on a GitHub issue. I have resolved it now.

This update appears to have broken a lot of plugins.

Sorry, bad wording on my part. By update issue I meant something where you had to have the version already installed prior to updating, vs. a fresh install (which worked for me).

I don't think ST3 broke anything. More likely Package Control 3.3.0, that was released three days ago, is the issue.

Immediately after the update if wasn't working, but restarting sublime seems to have fixed it

Good - he's working on Sublime. Bad - he's neglecting his plugins a bit. Hopefully with 3.0 out the door officially he can balance things out a bit.

The Sublime community is blessed to have his work supported in a way that lets him dedicate his time to Sublime Text as fully as he does. Given he one of few people who can work directly on Sublime itself, I am glad his focus is there.

Sublime is probably the first GUI text editor that left me satisfied. Everything else over the years was either non-portable, slow or lacking features. After two months of use I just had to buy it, it's worth the price.

- It's fast and responsive, can handle large files and personally I've never seen it leaking memory or crashing

- Love the goto anything and command palette, brilliant

- Text editing is great and can be easily enhanced with existing plugins

- Theming is great

- Project handling is a bit weird, but once you get the hang of it it's fantastic

I don't do use IDE-like features (auto-complete, linting, compiling, etc.) nor integration features (version control for example) so I can't say how it fares in that aspect.

Recently I tried VSCode and Atom. They are portable and feature rich out of the box, but God are they slow and memory hungry. I can definitely see the appeal for web devs but it doesn't work for me, I value responsiveness above all else (that's the main reason I avoid dedicated IDEs)

Have you ever seen UltraEdit? I dare you to try the trial, actually go through the exquisite configuration dialogs to make it yours, and then to ever look back :)

This isn't as a knock on ST, which I also like, and might even sometimes prefer for pure writing or temporary notes. But it doesn't even have printing out of the box, come on. For me that's too modern.. plugins are nice, but I guess I prefer UE to ST like I preferred old Opera to Firefox.

Could you mention what you're likely to be printing? What sort of uses do you have where you write notes in a plain text editor and print them out? I can't think of the last time I printed anything -- actually, I should say I can't think of the last time that I printed anything that wasn't a contract or HR paperwork type of stuff, which is never in plain text. I'm curious what else is still being printed these days. Also, they have plugins for that, if you're interested. You're correct that it's not built in, though.

I print code reasonably regularly! Makes going through it with a highlighter and red pen to get a sense of what it does and what should be changed quite nice, and it's good to get away from a screen every now and then.

I mostly do it when I'm examining modules in a codebase I'm unfamiliar with.

That's interesting to hear. (Since this is the internet, let me be clear: that's not sarcasm)

Theoretically I really like the idea of reviewing code on paper.

But typically, if I'm trying to familiarize myself with a codebase, we're talking dozens of files. Usually I'm going through three, four, five levels of code to see how a request is served up, and then my search will fan out sideways into even more files as I attempt to see what other processes in the application might affect the data I'm concerned with.

For example with songs I'm in the process of making, I like to have a list on me with just the titles so I can note down acoustic observations made while listening to them on various speakers or headphones, for the next iteration of "mastering". Or little bits of info for my wallet, e.g. when a handwritten thing gets so old that I want to refresh it, then I'd rather make a printout and stick that in there. I like to make "cheat sheets" too, for work stuff or for all sorts of things, which aren't elaborate enough to warrant desktop publishing efforts, just proportional text. Granted, I don't print a lot, but for me it's still some of the very rudimentary things one might do with text.

> Could you mention what you're likely to be printing?

To paper? Never. To PDF? I've printed to pdf to get nice looking formatted code for presentations.

I see print as more a brute force bazooka feature to get nice vector graphics out of apps.

IIRC there's a Sublime plugin (obviously) that supports copy pasting formatted code with syntax highlighting. I definitely remember using it at university. Printing to PDF to get syntax highlighting just sounds like an unnecessary step in between.

What I do:

* c/p in a gist

* when looking at the gist, print the page in your browser

* save to PDF

You may want to a look at https://github.com/mplewis/src2png (it was popular on here at the beginning of the week https://news.ycombinator.com/item?id=15220069)

I switched from UE to ST entirely because of UE's licensing/activation model.

Maybe it is different now, but UE used to require an internet connection for the registration... or you had to email your info for offline codes.

I have a bunch of isolated machines at work, physical and virtual, and got tired of dealing with UE and its registration requirements. Especially when ST gave a license key file I could copy around as needed.

So I switched. But yeah, I remember UE was a really good editor. Now I'm happy with ST.

I fell in love with UltraEdit about ten years ago, and then I forgot completely about it. It amazes me how "obscure" it is when it's a fantastic replacement for ST, VSCode, Atom, Brackets, and dare I say, even Emacs/Vim.

It always amazes me how Ultraedit is completely invisible on Hacker News. It's been a very decent option for a very long time.

>Have you ever seen UltraEdit?

A long time ago (>10 years) but then I mostly used IDEs for development. Didn't know it was still out there, I'll may give it another try, thanks

UltraEdit does column editing so much better than ST. I think I prefer ST overall but that one feature, when it's needed, is wonderful.

I use it in the office (on Windows) and yes, the column mode is the best I used. I did try the Mac version (of UE) but I think it was still in beta mode and I was not really impressed. But on Windows is really great.

I have and switched when I discovered how badly UltraEdit choked on very large files.

It's a very good editor; I always assumed it was windows only but it's multiplatform too.

First: I love ST. It is an amazing software that runs just as good on Mac/Linux/Win. In a pinch I will download and use it and never complain. Python extensibility is also great, since Python is the language I do best.

That said, and I hate to be this guy (not really), but... Emacs?

* Performance: Acceptable on modern day

* Goto Anything: Helm/Projectile, command palette = M-x

* Text Editing: It is what emacs does best as a program

* Themes: Yes

* Project Handling: Projectile

* IDE Features: These exist

* Source Control Integration: Many people use emacs purely because of Magit, it is a great git UI that people won't know about in general because it is runs on emacs

I copy my ~/.emacs.d to a server, literally my whole environment minus some tty nuisances just works. Everything. All my package versions. Emacs is stable enough that not many things break between big versions 23/24/25.

There are two areas ST wins: Out of the box configuration and performance on very big things. It is true the learning curve to settle all of these features I am mentioning is non-trivial, but Emacs is open source and, like I mentioned, very stable. It has gotten enough right in its initial concept that it will always be a fine environment to mangle up some text in files into whatever you want.

This brings me to my next point, there is nothing really new under the sun regarding text editing. Atom/VS Code/Eclipse/IntelliJ/Emacs -- you want to edit a bunch of files in a directory structure, navigate quickly and easily between files and get context important information about the code. Everything else is just flavors of those same chores, often specialized to a few programming languages or whatever new things are popular at the time (JavaScript / Can we build an editor in JavaScript).

I have seen very little innovation in the field of text editing. Light table was interesting, and its no surprise it is very Clojure/Lisp oriented. I am in a giant lisp machine when editing code in emacs, I can whack out a lisp expression anywhere and emacs will faithfully evaluate it for me.

Next point: If you work in a language or big project you need a way to navigate and generate lots of boiler plate in some languages (Java). IDEs help there, but...

Final point: The speed with which we write text as programmers is not the limiting factor of how much software we can produce and never has been.

My conclusion was to invest in one editor that has been around a long time and just not worry too much about it. I can use emacs in 20 years probably. Who knows about atom or vs code or whatever. So it isn't a pro emacs, you should use emacs reply to you, but that over the long run, using the same system to edit text is probably worth it. Stick with one thing and ignore everything else unless you are forced to use some IDE, and even then really try to avoid it. Even then I am pretty sure I could do my job with Gedit or Notepad without much of a productivity hit.

edit: formatting

> there is nothing really new under the sun regarding text editing

This may be true in for single heroic innovations, but sometimes the 'new' can emerge from successive refinement. I find this to be the case for IntelliJ. Working in it in static languages (at least the ones it knows well) just feels to me quite different from merely 'editing text'. The depth of code intelligence, combined with hundreds of small refinements in the way commands like refactoring work, affords a sense of working with syntax and structure, not 'text'.

> The speed with which we write text as programmers is not the limiting factor of how much software we can produce and never has been.

Again what IntelliJ adds for me isn't 'speed', but a cognitive sense of moving up the abstraction hierarchy. I'm thinking in terms of things like 'add a parameter to all call sites' much more than I am editing 'text'.

(caveat: I did use vim quite extensively many years ago, but never got beyond the learning shortcuts phase with emacs. It's possible either of the big 2 have similar refactoring and code intelligence packages available now)

> Working in it in static languages (at least the ones it knows well) just feels to me quite different from merely 'editing text'. The depth of code intelligence, combined with hundreds of small refinements in the way commands like refactoring work, affords a sense of working with syntax and structure, not 'text'.

These days, out-of-process, editor-agnostic code analysis engines or "language servers" provide many IDE-like features to any text editor that is scriptable and can do basic IPC. Examples include:

* Tern for Javascript

* rtags for C and C++ (I use this all day, every day, at work, it's remarkable how good it is)

* Go's oracle tool

* racer for Rust

* tools.nrepl for Clojure

All of these have solid editor support; at the very least, a good package is available for Vim, Emacs, and Sublime Text.

IDEs tend to be all or nothing. I'll open one on occasion for a particularly nasty debugging session. I wish, though, that half the development effort going into IDEs were instead directed towards better editor-agnostic tooling.

The only one of those I have tried is tern, which didn't come close to IntelliJ's analysis, but obviously Javascript's a harder beast in this respect than Java, kotlin, Objective-C etc.

> I wish, though, that half the development effort going into IDEs were instead directed towards better editor-agnostic tooling.

In principle I agree. I like the thought of composable tools. But I wonder in practise. I'm trying to picture the vast array of facilities integrated in IntelliJ (analyses, many refactorings, syntax-aware snippets, amazing code navigation, structural search & replace, data flow anaylysis, syntax-aware select, omnipresent code generation, etc) integrated into a text editor via external tools. I suppose it could be done, but it seems unwieldy. And with how much work (setup and maintenance) on the part of the user? How often would things clash or need fixing? And with how much discoverability (a critical feature of IntelliJ)?

What I do know is that working with code (not mere text!) in IntelliJ is quite a joy for me. For now, I'm glad someone has built it, even if the 'ultimate' composable set of tools might have some advantages.

Not everyone has this luxury, but I simply avoid languages that require an IDE to be productive or think abstractly. Alas, not everyone can get away with this. Also, IntelliJ is also a very nice editor. I have used it a decent amount and liked it well enough when editing Java.

I've used IntelliJ IDEA since forewer and I can only agree, and add a little bit of my own.

Having worked extensively with primareliu Java in IDEA, I have come to consider syntax something that is more a presentation layer issue than something actually important to the core differences between languages.

I suspect that this shift in what I consider important parts of a language is at least partially down to an IDE that for the most part lets me think about the concepts I am expressing, rather than their superficial expression as text on a computer screen.

I agree that finding an open source, well established and powerful editor is an investment worth making. That basically means vim or emacs. Once over the learning curve, you're setup for long term productivity gains and stability. Since neither editor is going anywhere and they run on pretty much anything.

But for many people, the initial learning curve just isn't worth it. Sublime, Atom and VS Code are all very popular because you just install them and start coding.

I'd love to see someone shake up the space with a "modern" take on emacs and vim: native, fast, powerful, very configurable, terminal based, but also with a conscious view on the initial learning curve and out of the box usefulness.

> But for many people, the initial learning curve just isn't worth it. Sublime, Atom and VS Code are all very popular because you just install them and start coding.

This does a great disservice to the quality and functionality of Sublime.

I was a fluent vim user before switching to Sublime and have never looked back. The smooth (i.e. not line-by-line) scrolling and powerful multi-cursors were enough to draw me. The little things like a beautiful file explorer sidebar with icons-per-filetype, great package manager you can launch, search, and install from right in the editor, and yes, occasionally even the mini-map, all keep me here.

There's a pervasive meme on HN that Sublime took off because of its lesser learning curve, and that if you're a professional programmer you should make the investment to learn vim and emacs. But this ignores that a lot of people who have already made that investment still find Sublime to be a better editor.

While I can understand the importance of using an open source editor, that particular argument doesn't sway me since I'm happy enough using Sublime for now. If it stops getting updates or whatever, I can always switch back to vim, but for now I will use what I feel is the best editor for the job.

Sublime took off cause it just works and it has a nice default color scheme and just enough batteries included to make it a functional programming environment for the 80% of programming situations. And it now has enough addons to increase quality of life in many areas and or solve the other 20% in many cases.

If I install emacs it is just okay. It probably has syntax highlighting, depending on the Linux distribution. It probably doesn't have any quality of life improvements for language X. I can drop in a theme and colorize it pretty quickly to my liking. From there, emacs is still very powerful, but there are a lot of things I really love for long and deep editing sessions when I am going through a lot of files on a code review or what not.

Don't forget: it can be arbitrarily extended in a popular programming language and it has an unrestricted free trial.

ST or Emacs?

ST. Python is the popular language. Elisp is... not popular. But you can do CL in emacs.

As an aside I found the Leo editor not too long ago. http://leoeditor.com/ -- one of the few editors that made me think "that is a little new". Leo is implemented in Python.

I wish there was a modern equivalent of emacs, only all Python. :)

Yuck. I prefer what Neovim did: editor comminicates using RPCs with arbitrary external programs using over a socket. Zero language lock-in and maximum extensibility.

Addendum: then again, you could use Python to do that.

> I'd love to see someone shake up the space with a "modern" take on emacs and vim: native, fast, powerful, very configurable, terminal based, but also with a conscious view on the initial learning curve and out of the box usefulness.

This sounds like the micro text editor. I've committed a few features to it and use it daily:


You might be interested in checking it out.

I think you should forget the whole "terminal based" thing. TUI text editors don't make much sense anymore. Native GUIs are much less clunky and much more discoverable.

I have my development machine at Digital Ocean. My local machine is an under powered macbook air (underpowered now that I use lots of docker for development). I'll probably just get a tiny chromebook next as I just want a dumb terminal.

I run emacs over ssh/tmux. I love it, emacs never shuts down, all my shells run through emacs as well. I can use any machine with my sshkey and be back to work immediately. I can increase the size of the machine, have public access to my dev box if needed. If I have a bad internet connection it doesnt matter since its just sending low bandwidth over ssh. All the real bandwidth comes from DigitalOcean and is super fast.

I do pretty much the same but using mosh instead of ssh. It's really stable even on low bandwidth of crappy mobile connection.

TUI still has advantages. Editing files over SSH is simpler, the app is easier to build cross platform, and for me personally the lack of a context switch when I'm in my terminal is nice. Even tiny little differences that keep you in your flow add up.

I wish people would learn to appreciate that UX is subjective. I find most GUI-based editors to be RSI inducing. The level of effort required to change that is not worth it for me. Others may disagree. And that's fine.

Just because it's a GUI doesn't mean you have to use the mouse. I find gvim much faster than vim in the terminal because it doesn't have to print a screen full of glyphs for each update.

On the other hand, vim + tmux is pretty awesome, and for that reason I prefer vim on the terminal. I've never particularly liked gvim, even if I'm not using tmux. In Linux, hitting the start or end of a buffer causes some sort of screen redraw that I've never been able to fix, but that might have to do with my configuration.

Well let's not generalize on the clunky part; there's always Visual Studio ;-)

I would generally agree with your point, especially on the discoverability front - this is why emacs and vim have an often dreaded learning curve. But OTOH, there are also a couple of rather nice benefits to the terminal focus that I don't currently see in more GUI focused editors. It would certainly be doable to replicate them, but it seems like no one's trying currently.

So I would hold off on forgetting about the terminal until there's a viable alternative around.

Oh, and: I've seen my fair share of botched remote access solutions on the Windows side of things and would probably use any TUI editor short of ed over any IDE in those situations. At least SSH is responsive.

I use Emacs in both modes.

Check out kakoune, TUI based and very discoverable.

But for many people, the initial learning curve just isn't worth it.

I don't believe that. The people who snub Vim and Emacs, but are willing to sink time into learning new editors, new languages, and new frameworks have no excuse. If they were less susceptible to marketing, then they'd be able to learn something old every once in a while.

I'd love to see someone shake up the space with a "modern" take on emacs and vim: native, fast, powerful, very configurable, terminal based, but also with a conscious view on the initial learning curve and out of the box usefulness.

People have been attempting this for years with little success. The only projects I know of that have traction are Spacemacs and Neovim. The former is just an Emacs distribution and the latter is backwards compatible with Vim.

It is in fact possible to learn and KNOW Emacs and Vim and still prefer alternatives to either of them.


Kakoune doesn't run on Windows, so not cross-platform.

Current Status - from the readme:

This is still a project in its early stages. The Mac build has basic editing functionality (it was used to write this README), but looks very spare and is still missing essentials such as auto-indent. At the moment, it’s expected that its main community will be developers interested in hacking on a text editor.

Yeah, I'd love someone to make a text editor that is:

* simple and non-tedious to use by default,

* can run via terminal and ssh,

* supports vim themes,

* full support for Language Server Protocol,

---* (Can work with clangd)

* fast and polished

Whereas Sublime "just works", emacs doesn't.

But when you're on Windows, boy you have to do a LOT to get it to just work. I tried it (and spacemacs, as I'm quite familiar with vim) last month for doing some python work on Windows... I just gave up after 2 days.

A couple of things that normally annoy me are that it's just a .zip file not an installer so the user has to figure out a "sensible" place to install it is ... and that there's like emacs.exe, emacsclient.exe, runemacs.exe or whatever - a bunch of .exes and it's not clear unless you've dug around a bit[0] which is "correct".

But after that it was relatively plain-sailing for me. What else was troubling you? Note: it's possible that we've got totally different workloads and needs - not saying you're doing anything wrong, I'm just curious

[0] - OK you don't have to dig far, (it's on https://www.gnu.org/software/emacs/download.html) but it still annoys me a tiny bit

>That said, and I hate to be this guy (not really), but... Emacs?

I intentionally didn't include console editors, totally different beasts. Never used emacs, only learned vim and only touched the surface, mostly use it when I ssh to remote machines. The pinnacle of flexibility and responsiveness but the learning curve... it's just too much for me.

You're right though, it's a great investment and as the years pass and mouse usage takes its toll on my right hand fingers, keyboard only editors look even more appealing. Whenever though I sit down and try to code in vim I find my productivity plummeting. I know once I ride the curve things will improve but unfortunately I find it hard to persevere

I use emacs in an X window 90% of the time, but it almost universally translates over to a tty. Fonts are just easier when it runs as an X window :)

> keyboard only editors look even more appealing

You don't say what editor or IDE you're currently using [ed: unless you mean ST? Which would be odd, because it doesn't need a mouse for anything], but few require mouse usage for much nowadays. You might be better off, at least in the first instance, just learning the keyboard shortcuts for the tool you're currently using.

ST is not a modal text editor. I know there is a vim emulation plugin for ST that's reasonably good, but at face value someone who has mastered vim can edit text faster than vanilla ST because of vim's modal and compositional nature.

Maybe grandparent poster just likes Sublime better. If he was an emacs user, would you have posted basically the same thing with "...and I hate to be this guy (not really), but... Sublime?"

Gonna take advantage of having an emacs wizard around -

I love vscode, but agreed, it's bog-slow sometimes, especially when I run in ubuntu. I'm looking around at emacs and I like what I'm seeing, but something that perhaps seems silly is still a bit of a turn off for me - it's pretty effing ugly. I don't mean in the theme sense, I mean in all the stuff cluttering the codepanes.

In VScode I usually ctrl+k z for "zen mode," where the only thing you see is the tabs of your open files, and the code. ctrl+shift+e opens my file explorer when I need it, ctrl+b puts it away (or ctrl+p to go straight to a filename). Same for terminal, ctrl+`, do a git thing, ctrl+` buh bye. Anything similar for emacs? That level of quick control over the UI?

You can easily do that in emacs.

Throw this in your init.el

(scroll-bar-mode -1)

(tool-bar-mode -1)

(menu-bar-mode -1)

(global-linum-mode 0)

And it will instantly look a lot cleaner. If you want to toggle it you can turn it into a function and bind it to a key.

> (scroll-bar-mode -1)

...what? You're gonna take away the scrollbar just like that? Implying that the scrollbar is worthless? Scrollbar is very important to me at least, (especially a well-implemented one). It gives me a quick and instant visual understanding of how big the file is, where I am currently in the file, etc. The only improvement I have seen on the scrollbar is Sublime's minimap... I knew there exist emacs methods to get it working, but I could not get it to work (on Windows) in less than a few days so that's of no use to me.

The wonderful thing about the list provided by your parent is that it's optional. Pick and choose the pieces you want. Those may not be the same pieces your parent wants. Fortunately both of you can choose without inconveniencing the other.

Worth noting also, with the default settings, Emacs tells you what your position is in the buffer in the mode line (Top, Bot, All, or a percent), so the scroll bar is redundant as a visual indicator.

There is two default UI elements in my emacs config:

Line numbers on the left and two lines at the bottom. One for the status bar and one for the "mini buffer" (where you put commands, search text, and receive short messages). Everything else is just the buffer (file) I am editing. You could turn the line numbers off with a simple configuration and key binding if you wanted, along with all the other UI elements.

This goes to the "Zen" of my emacs config. Its always just the buffer I am editing unless I need another dialog or UI element. The trick to making this work with no tree based folder explorers are Projectile and Helm. Projectile adds "project" concept to emacs and Helm adds fuzzy searching of everything almost everywhere that matters. Both can be taught to work around git repos. So you can say, whack some keys and see a list of files in your project. Then start fuzzy matching in the helm buffer that results by typing say three letters of a file. If its unique you press enter. File is open. If I really need a tree or explore a directory there is `dired` mode which lets you walk around the file system. There are even tree modes that I find pointless after using helm and projectile.

There is nothing built in by default. That is the biggest problem with emacs. You don't just sit down at it one day and get a polished environment that fits with what you like. It is something that is worked at. You eventually learn at least a little elisp and your customization powers keep growing. Eventually you either quit emacs or are good enough at lisp and configuring it that it becomes obvious, if not trivial, to do most things other editors can do.

Generally speaking it is almost certain you can customize it if you are asking a question like, "Can I do X.". Everything in emacs, even configuration dialogs are just text buffers. You just teach emacs how to interpret different text buffers in different ways to implement special functionality. Most of the heavy lifting of integration with most languages is already done if your language is remotely popular.

This was a long answer, but that goes to the heart of emacs. If you want a zen mode, emacs is there for you. You want a mode with weird little widgets and build output things everywhere, emacs has your back. :)

You can make your Emacs setup minimalistic by hiding the toolbar and scrollbar. This blog post gives a pretty good starting setup. https://emacs-doctor.com/emacs-strip-tease.html

C-x C-f opens files, but will also open a file explorer if the path is a directory. The 'q' key will immediately close the file explorer. I'm not sure about git interaction. I usually just use the Emacs shell to enter git commands.

Magit in emacs is highly recommended.

Just get emacs prelude and configure it to use helm everywhere. It is so close to my previous custom emacs setup that I just install prelude now on a new system and run with it.

I guess I never understood the custom emacs distributions for long time users. If you had a system that worked... lugging it around isn't hard. It is usually a few hundred lines of elisp plus packages, which you can freeze and zip up and they are probably going to work on an emacs release 5 or 6 years in the future.

I don't worry about it cause I just backup my .emacs.d and restore it anywhere I plan on editing very long. My whole .emacs.d with tons of modes and add ons is 73MB.

There /is/ something to be said with not making a mess of your initialization file and customizations. It is also very possible to keep glomming addons in and eventually get to a difficult place where various elements of extensions/modes/packages are fighting and that can be frustrating and I can see why many turn to well tuned emacs distributions.

Emacs gives me RSIs due to the awkward chording. I love the idea of Emacs, but the reality is too painful. I love Vim though.

> I copy my ~/.emacs.d to a server

I think if you work on servers a lot, then it's a major advantage to use something like emacs or vim, because your editing environment is 'just there'. But heaps of developers rarely, if ever, find themselves in a terminal, so the portability advantage just isn't there for them.

>That said, and I hate to be this guy (not really), but... Emacs?

The shortcuts make it a non-starter for me.

Having to configure stuff even more so.

Exactly. Many people want their editor to be very flexible and configurable and micro-optimizable to suit their particular needs, and that's awesome if it's what floats your boat. On the other hand, there are many of us who want to think about our editor as little as possible.

Sublime isn't perfect and it still requires tinkering from time to time, but its power vs. time investment equation is absolutely fantastic. It's a very get-shit-done style of editor. Emacs/vim are amazingly powerful but they are also big time sinks, both in terms of the initial learning curve and later customization.

You are right. I view emacs as a sort of religion or philosophy of computing. I don't use the metaphor lightly. It is pretty apt. It is my religion, so I would never put it on anyone else. I will share the cool stuff about it, but not in an annoying way.

The philosophy of using it is that eventually, in the long long term, the productivity gains should be reasonable or at least competitive with any other editor or environment. It should be a joy to use. Since I know I will probably only ever make a living crafting text I view it as an imperative to deeply master my primary tool. The other aspect is that it is free and old and stable. I can't ever achieve an RMS level of using open source things, but everywhere I can, I want the truly open thing, even if it isn't practical. I run Linux for this reason, even though it was a giant time sink making it behave and getting everything just so. But I can make things just so.

This fits in also with creating a distraction free environment. i3wm, emacs, I barely even see application menus these days, but I still use a GUI, I am not crazy. But it fits with my philosophy/religion of computing, so I am going to do it this way. If others agree, great. If they want to know about it, I will share. If they think it's crazy, from their perspective they may be right!

Learning curve can be a pain. Customization afterwards not necessarily, but I see where you're coming from. But once you are competent at Vim and you have it looking like you want it, it is nothing short of delightful. I feel motivated sometimes just because I get to work in Vim.


spacemacs is solid if you want to try out a well configured emacs. magit, org mode, helm are amazing tools and part of the killer feature list of emacs. magit is so good even if I stopped using emacs tomorrow I would still use it for my git gui.


I try such distros every two years or so to see if it's gotten any better. Usually it's not worth my time from the start (fun thing is, I used to use Emacs for a year or so in uni, before turning to Vim and never looking back -- until I had to do Java, which meant Eclipse. I got back to Vim, but the last 5 years it's all ST with some sporadic attempts to like VSC).

Let's do this in realtime. I downloaded the latest Spacemacs.

1) Opening. Hmm, it's a zip not a DMG. Already losing me, but I'll stick, I'm not that shallow.

2) Hmm, it's just a bunch of source files? I guess the prominent "Download" link just gives you the configuration, not a bundled editor. I have to get Emacs from somewhere else. Grr... Lemme check their "Quickstart".

3) Yeah... the Quickstart link just gives you some BS that assumes you already know how to install this, and the "Install" button just gives a git command line to clone their repo. Back to square 1.

4) Fuck it, I'll use brew to install Emacs. Spacemacs doesn't even say which version of Emacs it works best with. I'll risk with what brew gives me.

5) Brew installation ends fine, with a caveat: "Please try the Cask for a better-supported Cocoa version". Now you tell me?

6) Anyway, I have an emacs, lemme git clone spacemacs now. OK, comes up, nice logo. Now, I have a huge emacs window -- so why is it asking me "first time" config questions in a tiny space at the bottom? And why doesn't it explain better what the various options (distro flavors) entail?

7) Let it do its install, then play around a bit. Close the window in evil mode (q!). I open emacs again, greeted with this:

  Found 1 new package(s) to install...
  --> refreshing package archive: gnu... [3/3]
  --> installing package: evil-unimpaired@spacemacs-evil... [1/1]
  An error occurred while installing evil-unimpaired (error: (wrong-type-
  argument package-desc nil))
Oh, and:

    - Cannot find any of the specified fonts (Source Code Pro)! Font settings 
    may not be correct.      
8) I open some local source files to check. A buffer window appears on the right side with "warnings":

  Error (use-package): helm :config: Symbol’s value as 
  variable is void: helm-bookmark-map
Yeah, fuck that. Back to ST3.

To play devil's advocate here.

you clicked on download instead of install. now, to be fair, download shouldn't be on the main screen imo but still install is more in line with what you were looking to do so it should have been the thing you clicked on. Cloning the repo is installing it if you have emacs installed already.

It's open source so this can verbiage can be fixed. It's not a stand alone editor as it's a bunch of configurations to emacs.

https://github.com/syl20bnr/spacemacs has better onboarding docs than the homepage does.

At the end of the day it's a question of how much customization you want out of an editor. Reading the tone in your comment seems like you didn't want it to be successful at all anyway.

That's actually a pretty good barrier to entry. With emacs/spacemacs, if installing it is too frustrating for you, then configuring and using it is going to be way too frustrating for you.

Yes. And god forbid it plays out of the box end to end with configuration/elisp being totally optional.

My opinion is that emacs/spacemacs is more optimized for day to day work, than for the out of box experience. Personally, I'm fine with that, and there are some great editors (like sublime) that work out of the box if you prefer that.

However, the people generally working with/on emacs are not looking for that, they are looking for an editor optimized for customisation. They have already gone through the experience of setting it up, so why would they spend most of their time making it easier out of the box?

There is a cost to everything, and so far most people on emacs side have decided to spend their time on improving it's power, not it's ease of use.

For the most part it is completely optional to configure it and use elisp... it will ask you if you want to install a layer (group of packages) when you open up a filetype that it knows about and can handle. there is some configuration akin to setting up completion backend, linters, formatters etc that are applicable to your language/filetype of choice. The benefit of that is that the external tools are kept external where development of them can continue unimpeded by the development of the main text editor.

similarly, once you get to know the language quirks, it's a decent language that is basically a DSL for text editing. The power you get from being able to lookup how something was implemented by the text editor developers easily is extremely powerful.

If you like vim's idea of modal editing, emacs is 100% all about that. languages are major modes, minor modes are functions and keybindings that can be enabled and grouped together at will. transient buffers are used to send output to and are treated as normal buffers that you can do things to. The architecture and extensiblity of emacs is pretty amazing actually. Every key press is just a function and every key can be rebound based on current context using things like mode maps that get layered on top of everything and have precedence based on where you inserted the binding with sensible fallbacks when required.

Disclaimer: I love vim as a concept but hate viml and the plugins that were hampered by the lame programming language that was designed by bram.

if you have not tried magit and org mode you just don't really understand the power of emacs and the power of having everything be a function. spacemacs just makes setting up that powerful environment fairly painless but it's still a very old architecture with some warts.

but...but...my poor pinky finger.

Yes. I actually have a hard time using control now if it isn't capslock, heh. On other computers IT IS NOT UNCOMMON FOR ME TO END UP TYPING IN CAPS. That said, caps to control is a great quality of life improvement for anyone that uses CTRL for anything, but especially emacs users :)

Recently, after some corporate IT crap required handing over my laptop for a weekend, it came back with a post-it note attached that said "your caps lock key is broken, let us know if you want a new computer."

> That said, caps to control is a great quality of life improvement for anyone that uses CTRL for anything

I agree. I've been a vim user for over a decade, but a few years ago I decided to try mapping capslock to ctrl and have never looked back. It's a much more natural place for a key that gets used as often as ctrl, while capslock is a key that is rarely used (as a matter of fact, my keymap doesn't even have a capslock at all anymore, and I haven't missed it!). Anyone who uses ctrl more often than capslock ought to give it a try, but beware:


It's a real danger!

I'd recommend this even for non-emacs users who use terminals a lot. The tiny bit of extra comfort from ctrl-C in a shell alone is worth it. Add in tmux ctrl-B, or any readline shortcuts you use a lot, and you're onto a winner.

Ya, i did this a long time ago. All my computers have capsLock key mapped to CTRL

Same here. That's the original layout of UNIX keyboards, after all.

very cool, I didn't know that.

I prefer VSCode if the hardware can handle it, but otherwise Sublime is my goto. If Sublime ever had as good Git support as VSCode I would permanently switch back to Sublime

Honestly out of the box Git management for VSCode is really good. I'm surprised how quickly it got integrated to my regular workflow, and I'm a terminal gitter to boot.

The ability to see side by side diffs and which files/lines I've touched is what did it for me. I still use git CLI for squashing, rebasing etc, but basic staging and committing I do entirely in VSCode now.

I've been very satisfied with https://github.com/divmain/GitSavvy and GitGutter. Together they are how I do almost all of my git.

another vote for gitgutter. jumping to the next changed hunk in my currently open file is a life-changer.


If you want git support out of the box, please give me a thumbs up on this PR. https://github.com/sublimehq/Packages/pull/1126 If you can read sublime-syntax files, code review is appreciated as well.

With this package, you can set Sublime to be your git editor like this:

  $ cat ~/.gitconfig
    editor = subl -w

GitGutter and GitSavvy have been fantastic for my Git mileage -- have you tried them?

I prefer VSCode mostly for the intellisense/auto-complete.

As far as git is concerned, I use it from the command line an cannot understand why anyone would want to interact with git from VSCode or any other text editor.

I don't know if these editors support it, but being able to access git blame from editor is convenient, also the editor can highlight all the changes you've done since the last revision (especially useful to hunt down unnecessary whitespace changes).

I think Will Bond has a git plugin that's supposed to be pretty good. I don't like VCS plugins in my editor though, so I've never tried it.

Text editors shouldn't come with hardware requirements.

All software has hardware requirements.

Even ed has hardware requirements.

Have you ever tried TextMate? Since like me, you don't like "IDE-like features" I can imagine TextMate 2 could be a better fit than Sublime, especially in respect to file management. And a lot of Sublime bundles are compatible with TextMate (and the other way around of course, since these bundles originate from TM)

Sublime Text is among the best pieces of software I've used. I bought the license a couple of years ago, and would gladly pay the same amount again.

I love its core functionality: speed and stability, search, command palette, file navigation, Goto. I've used a number of editors over the years, and none have felt as fundamentally sound as Sublime Text.

The best feature of all is the Python plugin API. Sublime lacks some OOTB functionality of newer editors, but the plugin ecosystem makes this a non-issue if you're willing to invest in extending your editor. If you write code 6+ hours a day, you should be. Git integration (GitGutter and GitSavvy) is awesome, as is linting (SublimeLinter) and project management (ProjectManager).

I wrote an HTTP client plugin for Sublime called Requester (https://github.com/kylebebak/Requester) in a few thousand lines of code. It matches Postman, Insomnia, Paw et al. on features, and in my opinion handily beats them on usability. It's the plugin API that makes this possible.

Here's my guess on what Jon Skinner set out to do with Sublime Text: build a rock-solid text editor and make it as extensible as possible. Until someone does this better, Sublime is the best editor out there.

Sublime just makes editing fun for me. Especially with the ctrl+d multi-selection.

When it comes to editors, there's definitely a big spectrum of how much it does vs how lightweight/simple it is, and I think Sublime Text, for a majority of my usecases at least, is in the sweet spot.

Don't get me wrong, full fledged IDEs are useful sometimes too, but every time I go to use them, I always spend half my time searching and fighting the program instead of being productive. That's what I mean by Sublime making programming fun. Things don't get in your way and everything is very clean. Sure it may do less, but at least it doesn't drown you in features you don't use either.

I love Sublime for writing anything, not just code.

MarkdownEditing (https://github.com/SublimeText-Markdown/MarkdownEditing), optionally combined with Distraction Free mode, is my preferred setup for anything that's not code: docs, blog posts, prose, etc.

The plugin API seems powerful enough but last time I tried to play with it, I didn't get into the groove of things because docs were incomplete/not-up-to-speed/blog-posts-elsewhere-also-often-outdated, all this combined with Python-the-language it felt a bit too fickle/messy to me to get serious --- there was no comparison to the quite brilliant docs at https://code.visualstudio.com/docs/extensionAPI/overview when I weighed my options ---

But now that 3 is out of beta, I'll take another look and once again hope for an exhaustive, complete, up-to-date and comprehensive documentation of plugin development

I've found the docs to be complete but lacking in examples.

Examples often say more in 10 lines of code than a page of docs, but I didn't find this to be an issue. To write Requester, I cloned plugins with features I wanted to implement (e.g. GitSavvy) and used them to complement the docs.

I think the docs could be improved with examples of small useful plugins that touch on various areas of the API. This might be a good project for the community.

Just to clarify: the docs are rustic but complete. The API itself is immensely powerful.

Just checked out Requester today - what a fantastic plugin! We use Postman a lot but I can easily see this gaining traction in our office. Great job.

> I bought the license a couple of years ago, and would gladly pay the same amount again.

Same here... In fact, I just did plunk down my money again.

> and would gladly pay the same amount again.

Looks like I'm going to have to. Only free upgrades to people to bought after 2013...

Which is very generous. It's FOUR years ago!

That is when the BETA builds started, you effectively pre-ordered it if you bought a license at that point in time.

No, I bought a license for ST2. Also, the beta was opt-in.

>apt/yum/pacman repositories for Linux

Major kudos to these guys for distributing their software properly on Linux!

Let me rephrase that.

Major kudos to these guys for enduring distributing their software properly on Linux!

In the past few months, I see some popular linux softwares started to using AppImage (https://appimage.org/) to distribute prebuilt binary. Not sure if it is a good option for SublimeText.

I'm not a big fan of AppImage. It takes everything good about static linking and then fucks it all up.

That's what electro-build does by default. It's a terrible solution for end users, because it results in huge bloated statically-linked images, but at least it works everywhere.

I do prefer what Sublime Text is doing here though. Kudos to them, it's a lot of work.

It's easy to write a PKGBUILD for Arch, and IDK RPM but writing a Debian package is not that hard too, I've written a .deb that installs all the software I use on Ubuntu Gnome and it's just a directory and two or three files. Its docs could use a bit of help though, the Debian handbook is sometimes very superficial.

Well, offering a repo is yet another step further than just putting a .deb on the website.

Would you share how you made that deb? Sounds useful.

I just made a blog post on this: http://www.gkayaalp.com/deb.html


i think the hard part is maintaining it more than creating it. Library dependency and testing.

There is a lot more than just creating a package.

I vastly prefer it to hunting for executables on websites, thank you. And there isn't that many relevant distress to target, especially with things like Flatpack etc.

I second this. It's one of the rare non-default apps I'm installing on by Linux workstations!

I agree, too many just ship a deb or a simple archive, which is not the best update experience for non-Ubuntu users, even if i.e. the AUR solves most of the pain.

It reminds me that Atom still does have an official repository...

doesn't* --'

You can edit your post instead of replying with a correction.

Only within a 1h (or so?) window.

"One of the areas I'm especially proud of in Sublime Text 3 is performance". I totally agree with this statement. Writing software in sublime has certainly made my life easier. Kudos to Jon and team.

I opened 70mb file on a macbook air the other day, and apart from around 7 sec loading time, i didn't notice that it is any more than a 100 kb just by the UI feel. Really great job.

Every so often, I like to take a moment to appreciate how quickly fully highlighted files flit past while typing in the fuzzy file search prompt.

I like Sublime Text, but one thing that has really annoyed me is the lack of support for font ligatures.

See https://github.com/tonsky/FiraCode

It's a superficial reason, but I've gotten very used to them in Emacs, IntelliJ IDEA, VS Code and Atom, yes, I'm still using all of them interchangeably :-(

Supporting this is apparently requires major changes to their rendering engine so probably won't be added until at least version 3.1 or 4.

I find it as superficial as enjoying your editor font. It definitely matters to many of us I think. Sublime has a lot to be desired on the font rendering front from my perspective. I will definitely revisit Sublime once they tackle that presumably in 3.1 or 4.

Kudos to the ST team for never trying to patent or sue random companies for using the multi-select feature. That feature alone has helped improve so many editors.

Note that that feature was in TextMate, which was Sublime's inspiration.

No, it wasn't. I believe TextMate had a feature where you could edit multiple lines at once, but it was part of a column selection mode, it certainly wasn't free form multiple selections.

Multiple Selections were one of the earliest core features built for Sublime Text, and weren't inspired from other software. I don't think TextMate existed when I implemented this.

PC-Write had it, long, long ago and .. like ST .. this feature was the only reason I used that app.

I can't comment on how well it was implemented in PC-Write but it won't be far-fetched to say that Sublime made multiple-selections mainstream. I had seen some references to vim plugins that offered that feature but it had limited abilities. The three ways of doing multiple selections—selecting multiple instances of a word, placing multiple cursors downwards in succeeding lines, and using point and click—made Sublime's implementation vastly more powerful.

Back in the day (early-mid 80's), PC-Write was kind of like the SublimeText of the era.

Multiple-select was indeed one of the key reasons to use it, as a programmers editor, even though it was intended for more general-purpose word processing tasks.

Anecdotal experience: my first unboxing of ST2 led to the proclamation: "finally, someone has done multi-select cursors again, almost as good as PC-Write back in the day!"

hey! I used PC-write back in the nineties. I have nothing to add, just happy to find another user of that softeare.

You are correct - as you already know. I was wrong

TextMate 1 had multiple column editing, but true multi caret editing was not until TextMate 2 in 2011.

It had already been implemented in the alpha versions of TextMate 2 earlier. But it’s quite plausible that the two implementations were independent.

... Unlike most of Sublime Text which was a close copy of TM features except for the ability for users to edit customizations, making the Sublime community a giant leech that sucked all of the air out of TextMate bundle development without ever contributing anything back.

Sad history.

EDIT: Leaving the original comment below for posterity, but it looks like the TextMate article I found on this doesn't predate ST's multi-select, so I have no idea which is true. I'll just take the ST devs at their word :)

Interesting. TIL

It even looks like TextMate implemented the individual copy/paste buffer for each cursor correctly as well, something that VSCode had horribly wrong the last time I tried to use it.

Do you know if this only worked with the cmd+click cursors, or did it also support the cmd+d mode to select the same word in multiple places?

Regardless, kudos to TextMate as well :)

Didn't Emacs have something like this before ST?

Probably. I've used it in Emacs for quite a while, but I'm not sure where that falls in the history with the sublime text.

Indeed, they at the very least made it known / very popular to use. I realized JetBrains IDE's have support for it as well and it comes in really handy from time to time.

multi-select is by no means an ST invention.

As far as I'm aware it is. If you know of an implementation done before 2008 I'd love to know about it.

PC-Write had it in the 80's.


In LAPIS (1999), you could not only select multiple things, as it could guess what you meant to select based on the structure of the text: http://groups.csail.mit.edu/graphics/lapis/index.html

jEdit appears to have had multiple cursors since at least 2005, though I don't know how useful they were.

Out of curiosity: I looked up the jEdit developer and it turns out that he's currently working on the Swift compiler at Apple!


They worked great and I stayed with jEdit long past when it was reasonable to do so because of that and so many other amazing features.

Why did it stop being reasonable?

jEdit was a seriously under rated text editor, I used the CRAP out of it in college.

Was? It is still being developed, the last version was released six months ago:


nice, I had no idea.

OS X has always had it, albeit a fairly limited implementation. Try it in TextEdit.

yes! this was the feature that won me over (on top of the great syntax highlighting on black background). /happy sublime text user

I honestly can't believe I'm saying this, but: can you please enable me to buy a new license for 4.0 even though it may not even be on the road map yet? Or switch to / enable a subscriber model which is paid yearly and gives access to all upgrades?

I rely so much on sublime for my day to day work and I fear the $80 or whatever I paid for it whenever ago is too cheap for the amount of value I'm getting out of it, and I'd hate to see this magnificent piece of software fall by the wayside because of an unsustainable business model.

Of course, if the business is perfectly sustainable then you know, carry on as you where.

I understand your appreciation for the software but begging for another subscription fee? One feature here is that the authors charge a normal fee and are not greedy in your pockets, like 1Password and many others.

I really do appreciate the classic software pricing.

It could be a higher fee, or more frequent but I'm really tired of all the Recurring costs!

Thing is, software unlike say a book or a movie has sometimes significant maintenance costs. There are always bugs to be fixed, security vulnerabilities are found, third-party libraries need to be upgraded and so on.

The danger of the one-off model is that it sometimes creates the wrong incentives. Features and even bug fixes get pushed into newer versions, which may even delay their release until there's enough of them to justify a new version.

A good example of this going awry is the RAW file processing for Photoshop (pre-subscription). Newer cameras would get added to new versions of the Camera Raw plugin but you needed a new version of Photoshop to use the newer Camera Raw versions.

I honestly don't mind the subscription model and I think Jetbrains has about the best one of all: you get constant acess to upgrades. When you choose to stop subscribing, that simply locks in the latest version you can use (ie you don't lose access entirely).

The problem is many companies price subscriptions wrong. They say "our software costs $600 so we should charge $50/month". Because, you know, recouping that cost in a year is somehow reasonable.

Well, in contrast to a physical thing selling two copies is exactly as costly as selling one. I'd argue that that more than covers the cost of continued development.

I can't say I recognize that wrong incentives in real life and I've yet to find a single piece of software that I'm willing to pay a subscription for.

It all comes down to business ethics, are you trying to rip off your customers? Most can't afford to piss off their paying users so they play nice (and I hope that is not the only reason they are being reasonable).

"Well, in contrast to a physical thing selling two copies is exactly as costly as selling one."

Statements like this ignore the costs of creating it in the first place.

No, they don't. Even if the costs of creation were extremely high, selling more copies wouldn't change these costs.

I happily subscribe to bunches of services. Some I don't even use, I'm just keeping the option open. Am I being ripped off? I don't think so.

Not everyone feels that way. Some services are worth it of course, but there are a lot of pretenders that have entered the picture who do not sell things worthy of a subscription. There is an older model of software as eternal, you finish the program and then move on to the next bigger and better program. At least then your old userbase doesn't get as shafted.

> not worthy [of your money]

It's easier to cancel a subscription than get a refund on a license.

But those are entirely different measures. Not worthy of a subscription does not mean not worthy of my money, it simply means it is not worthy of my money on a recurring basis. Perhaps I am one of those seemingly rare consumers that did not mind high one-off costs, even in my poorer years -- in the end you still own the thing, or at least it's bits. Subscriptions net nothing over the long term and their recurring spikes will eat people to death.

Well, I didn't mean to say that people buying subscriptions are being ripped off.

But rather that companies that don't offer subscriptions can't game on features and bug-fixes just to fish for upgrade fees before people feel (rightly so) ripped off. There is an implicit agreement on the support provided and if the company fails to deliver that people won't play along.

Isn't that the other way around? If I buy a license and I find a bug a year later, there's no guarantee of a fix. If I am subscribing, I can threaten to cancel.

It was a direct response to the parent.

Also, if you loose access when you stop subscribing that is actually much harder to do compared to not paying for an upgrade if you "own" a license of the software.

Good point. I guess your stance will depend on how troublesome the bug or feature request is.

You're merely ripping yourself off sounds like :)

I completely agree with you. I was a user of Balsamiq mock-up tool. But since they moved on to subscription model I stopped using it.

Balsamic mockups desktop is still a single 99 USD purchase. Maybe you thinking of the other versions?


Software is something that needs to be continuously developed and maintained. Charging a one-time fee is a fundamentally unsustainable approach. We all have to face reality and start paying for software as a subscription.

Now, that will likely end the bubble of "I can buy 100 apps, most of which I'll never use", which we have been in for the last 20 years or so. But perhaps the few apps that people will pay for will become better as a result. I'm hoping for it.

> Software is something that needs to be continuously developed and maintained.

Not always. Sometimes software is "done" and warrants no further changes or maintenance. I have run into this issue with mobile apps that worked perfectly to my satisfaction until the authors decided to "redesign" or "improve" it. I think subscription is a valid use case for customers who want active maintenance, but it should be opt-in. I detest the forced-subscription model that JetBrains attempted ("Didn't pay the latest subscription? We'll brick the application you've installed and are running on your hardware").

" I detest the forced-subscription model that JetBrains attempted ("Didn't pay the latest subscription? We'll brick the application you've installed and are running on your hardware")."

They never attempted that. You were always able to use the version you paid for, even if you stopped your subscription.

> They never attempted that

Oh, but they did[1]. You might have missed the initial drama as it transpired because of their quick U-turn (to their credit). I have reason to doubt the subscription plans they published which clearly stated that failure to keep up with the subscription would lead to bricked IDEs. JetBrains only backtracked[2] after receiving sustained pushback. There were HN posts for the initial bricking plan[3] and the backtrack[4].

1. http://bytecrafter.blogspot.com/2015/09/how-jetbrains-lost-y...

2. https://blog.jetbrains.com/blog/2015/09/18/final-update-on-t...

3. https://news.ycombinator.com/item?id=10170089

4. https://news.ycombinator.com/item?id=10278285

Sometimes, sure. But nowadays software is more often than not something that isn't simply done and needs continuous updates, in the face of security vulnerabilities, changing requirements (people using the software for changing use cases), platform updates (new version of macOS or Linux that breaks compatibility), etc.

Do they actually brick it? I thought they just stopped upgrading it.

Reposting links I added in another reply. Yes they announced the bricking plans (http://bytecrafter.blogspot.com/2015/09/how-jetbrains-lost-y...) and then backtracked (https://blog.jetbrains.com/blog/2015/09/18/final-update-on-t...) - which is why I said "attempted".

Quite honestly, I don't mind paying a subscription for updates. But I detest having to pay a subscription when all I want is a single copy and don't require rolling updates and support.

Charging a one time fee works just fine for lots of folks out there making software. It's especially good for anyone who's doing it in their spare time as they don't have to provide updates for every little niggle someone may find. Some offer a subscription alongside the one-off model, which is the best of both worlds if the developer(s) are continuously working on it. Others ask you to purchase the major upgrades but give minor upgrades for free.

There are lots of payment models out there, and the developers should pick what works for both them and their customers. There's no way most of my colleagues who have a license would have purchased sublime if it was a rolling cost, so that model seems to be working in their favor right now, at least for single developer purchases.

The public does not give two shats what it costs you. In thier eyes they want it faster cheaper better.

For a piece of software that enables your career, helps you to provide for yourself and/or family, that you probably utilize for thousands of hours over a given year?

ST3 is continuously improved. They are developers. They eat. They have families. Just like you receive a salary, because the software you work on, probably is never done. Why would anyone feel entitled to improvements for a flat rate? Is that not greedy?

Gratuitous appeal to emotion. Sublime is already paid software, they are already eating, the commenter wasn't asking for it for free. It is not on the cheap end either - I have paid less for way more powerful and man-hours-consuming image/vector editors. Everything is fine.

Well. As a long-time Sublime Text user who sort of drifted off to Atom and VS Code in recent times, here's the thing: ST really hasn't been continuously improved; the sole developer frequently goes incommunicado for months, and there have been long stretches when there weren't any builds released. There are certainly substantial improvements between ST2 and ST3, but it's been over five years between ST2's initial release and ST3's. This is--kind of a worrisome pace of development.

So, this leaves me of two minds about a hypothetical subscription model. If it meant that ST's developers had an ongoing income that allowed them to work full-time on the project, set a regular release schedule, and be more open with communication, it would be worth it. But if users had been asked to pay, say, $49 a year for the last five years with ST3 on an "um, they pop up once a year to say they're still working on it, I guess" schedule, I'm pretty confident it would have gone over even more poorly than what actually happened.

Yes. There are endless text editors doing the same for free.

I could never find one that is as well thought and as well written as Sublime, and I've tried many, many others before.

Switched from Sublime to VSCode a few months ago and I've found it better to my needs. It's free

I love Code. It's one of Microsoft's greatest masterpieces.

Not quite free – you're paying in ram. EDIT: I mean more ram than you would use in Sublime.

You already paid by buying a faster computer. Some don't want to upgrade their computer; in this case, sublime is well worth its price.

Well-written? I thought it was closed-source.

I assume he was talking about the lack of significant bugs.

> [...] that you probably utilize for thousands of hours over a given year

Assuming 4hs net per day, 20 days/month, 12 months .. thats 960 hours/year

And that's assuming that you spent all those 4 hours _entirely_ on your text editor; no cli, no browser, no nothing.. only text editor.

I'd say that on avg we should be doing ~200 hours per year.. Are there any study about this?

The average number of hours worked by a fulltime employee is approximately 2000.

Obviously not everyone will be using their editor more than half the time, but many do. And many often use their editor outside of work too.

Are you suggesting that everyone with an 8-5 job works in Sublime Text for at most 4 hours of the day?

I have a family and I have to eat, so give me your money and I'll give you a copy of VSCode that I worked on. I'll also give you the source code with that.

It's a way better deal (and arguably a better tool) than what you get with Sublime.

>For a piece of software that enables your career, helps you to provide for yourself and/or family, that you probably utilize for thousands of hours over a given year?

Speak for yourself.

He obviously speaks for people who use ST as their editor. The grandparent expressed his opinion as one, and the parent answered as another.

People who don't have no says whatsoever in this particular subthread discussion.

The first line is the part I'm quoting, which is obviously what I'm addressing. He's making assumptions about the commenter that are completely unwarranted.

So no, he doesn't speak for the people who use ST.

It's Hacker News. We're discussing a mainstream code editor. There are multiple discussions of code/libraries/engineering every day.

The generalization that implies that the OP is a person who codes for a living is far from "completely unwarranted".

There are plenty of students and hobbyists here. So again, take it easy on the assumptions.

If the description of the parent doesn't fit you then maybe you weren't his/hers intended audience?

No reason to tell everyone about it.

If someone comments: "It's not much to pay $70 for something you use professionally, and get value out of it everyday etc" -- it only adds noise to chime in "I don't use it professionally" or "I don't use it everyday".

It's a conditional sentence (even if implicit). If one is not on that category, then yes, obviously the argument doesn't apply to them.

Exactly, for every one person like this there are probably many who would not renew if they went subscription. It's not like they rely entirely on new customers, the company only has two employees and I'm sure ST3 will be out for a few years and then ST4 will come along. Software companies existed long before "cloud licensing" became the norm.

I am not fan of replacing classic pricing model with recurring payments but instead of forcing recurring payment, allow that option for extra feature. Make the base app as it is now and add something on top of that for small fee. Like I suggested earlier, make small cloud service for syncing configs, addons and projects between installations. It's not a big thing, not deal breaking, it can be even done using rsync by user but I would happily pay couple dollars for that functionality and I would support devs. Development is recurring cost. Without aggressive marketing, like for example Microsoft, it's very hard to pay developers from selling not recurring licences every 3 or 4 years.

I would be okay with them having an optional cloud service for syncing settings.

But the issue you talk about is where major version upgrades come in. Sublime text could cease all development tomorrow and not pay another cent towards development, should people have to continue to pay to keep using it? If new major feature are requested then that is grounds for charging and upgrade fee, otherwise I don't see why people should pay.

Greedy? 1Password requires servers. You’re “consuming” that every month. Perhaps if 1Password cost $200 one time?

1Password works (or worked) without use of their servers by storing datastore on 3rd party cloud, which incurred no costs to them.

5/month to store 1kb of passwords

No. $5/mo to manage my team's sharing of accounts.

I’m having a big problem paying monthly for a password manager . May have to just roll my own in swift

There's always KeePassX if they're still around.

There is no open source mobile alternative active or big enough to be properly audited/trusted.

KeeWeb.info - there you go. Open Source, $0.

I was using MiniKeePass but it was hard to keep sync.

I'm worried about using the hosted version of KeeWeb, also it doesn't seems as practical as a standalone app. But I'll definitely try to host myself (or use locally on the phone?) and give a try. Thanks.

Take a look at Enpass (https://www.enpass.io) it has a free desktop app and one off fee for mobile apps. I've been using it for tree or four months and it's been great.

Noone's stopping you from buying more licences, if you really want to support the dev. I think it's overpriced as it is (as another commenter mentioned, considering the competition is FOSS), but I might buy it one day, when I have more disposable money (assuming I won't have switched to VSCode or $hotNewThing by then). I really do like Sublime, and I'm happy 3 is out of beta, but seeing the userbase migrate to Atom/VSCode is worrisome, and makes me doubt $80 is worthwhile investment (I'm mostly worried about third party plugin devs, which was once Sublime's biggest strength).

I use Sublime and paid for it a few years ago. That said, I also use VS Code which I really like for certain things. I actually use both of them daily and simply get different utility from them. If Sublime went to subscription then I would make do with other products. VS Code has a robust plugin ecosystem and I see that growing in the near future. I honestly think Sublime will eventually end up as FOSS. Whether that is from a drop or increase in subscribers I don't know.

It needs to provide you with just enough productivity to help you work for 2 extra hours, and it'll have paid for itself. $80 is peanuts for this.

I don't live in the United States so my purchasing power from two hours of work is exponentially smaller, but I get your point. Truth is, I don't really use Sublime (or any other text editor) that often anymore, because I'm working with Java and using IntelliJ IDEA for that. If I used Sublime Text professionally, I (or my employer) would have bought the licence in a heartbeat.

Does every developer earns $80 for two hours? No.

Can someone please explain to me why $80 for Sublime Text is a good price for a text editor?

I'm not flaming, I genuinely want to know. I've tried Gedit, Atom, VS Code, Notepad++ and I fail to see what Sublime offers over those to justify the price.

>I'm not flaming, I genuinely want to know. I've tried Gedit, Atom, VS Code, Notepad++ and I fail to see what Sublime offers over those to justify the price.

It's better than Gedit in every way, it's way faster and more lightweight than Atom and VS Code, and it has a far better plugin ecosystem and functionality than Notepad++.

I'd pay triple the price.

> it's way faster and more lightweight

Now you've got me wondering why editor speed is actually a thing.

Not only does the editor need to track project folders and open files, even something seemingly simple like syntax highlighting can be surprisingly complex[0]. The editor needs to be able to read your inputs (mouse and keyboard) and represent those changes onscreen just like every other piece of software. It needs to repaint the screen with your changes and, depending on the complexity of the editor, load and implement a host of features (auto-complete? bracket matching? code indentation? code standards a la pep8? updates to source control?), with every single keypress and ideally in a matter of milliseconds. All of that takes memory and different editors make different sacrifices.

Of course the extent to which this happens is the whole reason there are different editors. If all you want to do is write text with zero features, then speed isn't an issue at all and you can just write directly in your terminal to a single file with pico or something.

[0] https://code.visualstudio.com/blogs/2017/02/08/syntax-highli...

> even something seemingly simple like syntax highlighting can be surprisingly complex

I was troubleshooting something with a colleague who was using Atom, and every time he opened a new file, it'd take somewhere between 1 and 2 seconds before the syntax highlighting was applied. As someone who's in and out of files all day, that'd drive me bananas.

Slow software is frustrating. A text editor is a tool, it should help and not hinder.

There are a lot of factors that play into the speed (perceived and real) of an editor. Sublime is written in C, which is generally recognized as a performant language when used correctly. In contrast, Atom and VS Code are Electron[1] apps, which mean they have to deal with browser optimization. They both use a lot of tricks[2] to operate as quickly as they do. And in all editors, one bad extension can kill performance, from my recent experience Resharper in Visual Studio increases startup time by 3x.

An editor is not as simple as displaying lines of text a la Notepad, different editors optimize for different use cases.

1. https://electron.atom.io/

2. http://blog.atom.io/2017/06/22/a-new-approach-to-text-render...

Sublime Text is written is C++ FYI.

The same reason its annoying when any other kind of program is slow, especially interactive ones.

I don't want to see freezes for 2-3 seconds for some GC run, I don't want the letters to appear several 100ms after I've typed, I don't want to wait minutes for it to parse a large-ish file or update the autocomplete suggestions.


You mean why some editors are slow or why I care?

Not sure about the first.

I care about speed because I like to use "large" files. I prefer a few large files to many small files.

Try using something like xcode or a jetbrains IDE with a large codebase (for the record I like jetbrains software, but it can be slow comparatively).

edit: not sure I correctly understood the parent

Having tried Atom, it's the main reason I stuck with Sublime. Somehow, even typing was slow in some larger files...

It matters for huge files

So not source code, for the most part.

I don't use Sublime Text as an IDE (I have Intellij for that), mainly to edit scripts, large CSV files, as a scratchpad etc. but It's always open and increases my productivity a lot.

It's highly likely that OP simply stumbled upon it, or someone recommended it, and now he learned it and is used to it and knows how productive it can be.

Thing is, same is true for any other good IDE, of which there are plenty.

Personally, it always blows my mind that people don't use VIM :) if they actually spent 5-20hours learning VIM correctly they would not touch these inferior pretty looking UIs

Agree! I do almost everything in VIM for file-efficient languages and environments (NodeJS projects, Python projects, HTML, C++ ROS packages, most things that promote DRY coding, etc.)

For file-inefficient environments (e.g. Java/Android -- 79 files for a "hello world" Android app!) I have to grudgingly use an IDE because I have no idea how to update the 78 other manifest/gradle/workspace/resource/etc. files based on the 1 file I change and the IDE takes care of the magic, as much as I hate that way of working.

You can have your cake and eat it too.

There are some really great vim plugins for Sublime Text, including one that uses NeoVim. For me, Sublime Text is basically Vim with an nice UI and some IDE features built in. And Goto Anything. I wouldn't want to live without Goto Anything.

I don't want to put anyone off they're favourite editor but you can have all of that cake (well maybe not all the eye candy) in Vim/NeoVim.

e.g. Goto anything you can use the Ctrl-P plugin.

They're always inferior editing experiences, these vim bolt-on plugins. Missing key features like 'jk' mash escaping and other remapping support.

      { "keys": ["j", "k"],       "args": {"key": "<esc>"},    "context": [{ "key": "setting.actual_intercept", "operand": true }], "command": "actual_keypress" },
That, in your Sublime keymap, should map jk to escape if you're using the ActualVim plugin.

On the backend, it's using NeoVim... So it's literally just vim. You can even use vim plugins. There really shouldn't be any missing features.

Oh! Didn't know this. Thanks.

Whereas I view vim as simply a necessary evil and I mean "necessary" in the sense that I want to edit something over ssh.

Bret Viktor's oft-quoted "Inventing on Principle" talk touches on this subject where a lot of research has shown that bimodal editors are worse. While I can't argue for how true that statement is, it's certainly been my experience. I always find it jarring to think I'm in command mode but I'm actually in edit mode or vice versa. Any vim user has had the experience of pasting text in command mode. I just absolutely hate bimodal editors.

Additionally, staunch vim (and emacs) users I know seem to spend an inordinate amount of time screwing around with plugins to get some pale version of behaviour you get out the box on any halfway decent IDE or even text editor.

I feel the same way about learning yet another series of keyboard shortcuts for, say, tmux/screen.

You could argue the mouse is slower. I'm not sure if this actually holds up to empirical measurement (or at least not to the degree that's claimed) but the point is, the barrier for entry is so much lower. You can just click around and find things. If you find yourself doing something a lot, you can find out what the keyboard shortcut is.

Thus you don't need to spend 50 hours upfront on a GUI editor.

It's possible to go overkill with learning tons of unnecessary shortcuts, but in my opinion all you really need for tmux is attach, detach, split horizontal, split vertical, fullscreen.

5 shortcuts is a paltry investment for a pretty sweet gain.

And unlike an IDE, this can be readily available on any machine you SSH into. Which is also what makes VIM so handy.

From the point of view of "onboarding" (aka barriers to entry), as you say, point-and-click is miles ahead of the unfriendly-at-first-glance Vim.

But I'd like to say that I cannot honestly recall an instance where I was in normal mode and ended up pasting something -- I assume you mean Ctrl-V-ing and having the paste interpreted as commands? -- once I got used to it after leaving Sublime Text, which I'd say took me a couple of weeks of hating myself. It's a matter of how much effort one thinks it makes sense to put into learning the editor, but IME there is a significant payoff, even if it is hard-won.

I'm not sure if it's harder for some people to internalize (which is by no means a value judgment; it's why Emacs/ST/Atom/whatever exist and are powerful) but once that is done, one never gets confused about what mode you're in simply because one is almost never in insert mode! "Probabilistically" speaking, you are in insert mode for 0 time. ;)

In any case, I don't hate non-bimodal editors, just hope that the subset of those using them who have not seen Vim for what it is yet do so quickly. (And, of course, there are those who are genuinely repulsed by modal editing, and that's fine!)

I can't speak to "inordinate amount of time screwing around with plugins": I have a plugin for each language I use, and perhaps four or five others. The rest of my init.vim (I use Neovim) is about ... 30 lines long? It's my experience that most (Neo)vi(m) users use little of the power of plain vi -- I know that because I myself was like that for almost three or four years after I started using Vim. (This is ignoring, e.g. large Java projects, where IDE-style code navigation is basically a requirement if you want to get anything done.) But there is definitely a culture where a customized-to-the-hilt editor is seen as something to be proud of, and why shouldn't it? It's what you "live in", isn't it? :)

The 'micro' text editor is a fantastic vim replacement for SSH editing. It's a text-ui like vim, but it has mouse support (for selecting text, etc), and keyboard shortcuts that match bio stands, ctrl-s for save, ctrl-q for quit, etc.

Worth noting, Vim has full mouse support for scrolling, selecting text, resizing windows, etc. in the terminal and when working over SSH.

> You could argue the mouse is slower. I'm not sure if this actually holds up to empirical measurement

It is slower, and does hold up to empirical measurement. Yes, if you want 'ease of onboarding', then vim is not for you. It's a powerful tool for professionals, and like a lot of such tools, it requires some training. No-one ever complains that Photoshop requires training, yet a brand new user to Photoshop is completely befuddled and it takes a long time before they're proficient.

I've owned Photoshop since it was distributed on diskettes and I'm still befuddled every time I need its impressive power. (I'm not a professional graphics/photography person so I use the program only a few times per year now.)

I think this is a really good analogy to using pro-level text editing tools (e.g. Emacs or Vi). A mouse is great for poking around menus and operating sliders in a new or unfamiliar program, but Emacs and especially Vim are so fast to use without a mouse that it pains me to watch other programmers editing while using one.

ST is a fine editor, and I use IntelliJ for Java, but I've already spent the time (years) learning Emacs. Is it worth it for me to start over? No. I use IntelliJ for Java programming, but I use only its basic Emacs keybindings and the mouse for everything else (befuddled like I am in Photoshop).

You don't know the right vim / emacs users then :)

Counter-example: I used vim for years, then switched to Sublime.

I bought a license simply beacuse Sublime Text was the only editor that could open a wikipedia json data dump. I installed many editors before I more than willingly paid for Sublime.

For me, same thing: no other editor gives me the decent UI of a GUI-based editor, but with the speed of vim when working with large text files (e.g. 10MB+).

I will often use sed or other CLI tools to parse through giant files (1GB+), but sublime only takes a short bit to load in a pretty hefty file, or a do regex search across entire projects with multiple-GB of files, and it works natively on Mac and Windows, so even when I'm forced to work on something on my Windows 10 laptop, I can be productive.

Well worth whatever I paid for it 4 years ago.

It has almost all of the features of the Electron based editors (aka Atom) but with a C++ backend instead of HTML/JS toilet water. So you have all the customization you could ever want with the speed of low level code.

It's a dead simple editor, comparing to VIM or Emacs (both I used a lot). Everything behaves just as you would expect.

And it's fast, it's small, and it's got a lot of packages. What more do you want?

Terminal support? To be fair, sublime text is great and I use it sometimes, but if your job requires you to work over ssh then vim definitely beats all modern editors.

Yeah, I agree. That's when I use vim/emacs.

Speed and stability, search, command palette, file navigation, Goto.

And the excellent Python plugin API, which has made the plugin ecosystem so vibrant. Here's one I wrote, https://github.com/kylebebak/Requester, an HTTP client that goes toe to toe with Paw and Postman on features and outdoes them on usability.

I couldn't have have written something like this for any other editor, or any other platform.

I think it is in a similar situation as JavaScript. It is not strictly better than its competitors, but it still manages to attract an entire community of people[1] to contribute actively to the project and its popularity.

[1] https://packagecontrol.io/stats

Nothing is faster for me when opening gargantuan text files. It's never broken for me and never lagged either. I think that's pretty impressive, not to mention the customizability options.

Define gargantuan. I just opened a 900MB CSV and it choked part-way through and eventually opened after about a minute. Windows 10, 8GB RAM and running on an SSD.

Not to mention if you hit ctrl-f and search for something in the file. Another lock-up for a few minutes while it tries to figure out highlighting. This is something I've reported to them. I have never considered Sublime to be a good handler of large files.

Opened a 1GB text file on MacOS 15gb ram and SSD.

> Windows

Same problem on my mac with 16GB RAM and also an SSD

Agree, yet even SlickEdit - which costs several times more, is still running strong. I wonder what kind of audience pays for it.

The small things are really well done.

It's an organic, artisanal text editor lovingly hand-carved from gap-buffer wood by craftsmen from a remote Tibetan village. A steal at thrice the price!

What about the idea to sell 'silver/gold/diamond supporter' licences for 100/200/300$ to people who _really_ love your software?

I am not even talking about adding some special features, just enable people to pay more than the normal price and perhaps just add a nice 'Thanks for your extra support' badge to the 'Info' menu.

Buy licenses and donate to your local high school.

My local high school will refuse them. The computer science classes use Geany because it “builds character” and anything that looks remotely like an IDE apparently “does too much work for students”.

This is great, I will have my accountant look into how I can do this. Thanks for the suggestion!

Excellent, excellent suggestion.

Or give the diamond supporters the ability to vote on features. That may not be something Jon and Will would want though.

That is exactly what I would avoid. Don't add any special features to it (vote for features, faster support responses) because that will take time from normal development. Just clearly state: "This software cost x$. And if you love it so much that you would pay double the price then you are free to do so."

Totally. I bought my license in 2013 and just realized that I didn't even have to pay the 3.0 license upgrade fee! I have used Sublime daily for almost 5 years now and I almost feel guilty for paying so little. I'd be more than happy to pay for a future 4.0 upgrade fee now.

Bought mine November 2012. Just paid my US$11 now. Far too cheap; but... good on you Jon! Thanks for all the effort - you rock!

Same here. I honestly didn't expect this, that's a bargain.

As a former paying user of Sublime, I find this interesting, because their pricing seems to be right on target all around.

If there was a subscription available that offered some amount of extra features, I probably wouldn't have purchased anything at all.

So many nasty replies. :o(

(Some pretty great replies too – thanks!)

I'm not a ST dev, I have no financial interests in ST becoming more expensive or changing to a (for the company) more lucrative business model. I have tried alternative editors, anything from vi and emacs to VSCode and Atom, but nothing hits the marks for me as ST does. It has – in my opinion – great UX, its speed and efficiency is unmatched in the GUI editor space, and my favorite feature: it has rock solid buffer persistence.

I can't tell how many times ST has saved me from system crashes and just me being an idiot and restarting everything before saving anything, or removing a file but ST keeping a buffer. I've come to rely on this, and it's rock solid. Not once in the years since I started using it (I believe I bought my first license in 2012) have I encountered a situation where ST lost my data. I've never experienced this with any other editor (I'm sure vi and emacs can do this, but in years of trying, I can't come to grips with their UX – they're just not for me) and this feature alone has paid for the price of admission for me, many many times over.

In short – I don't care that there are FOSS alternatives. I don't care that there are alternatives that are "better" by some measure. Until they can match these three things, in increasing order of importance, they will never compete with ST for me:

- Great UX

- Speed and efficieny

- Rock solid buffer persistance, even for unsaved files

Someone in this thread suggested I buy licenses and donate to a local high school. I'm going to seriously look into this as a viable way to both support the company producing ST, as well as hopefully enabling kids to learn programming, or write novels, or whatever they want to do with the software.

Agreed. I am a very appreciative Sublime user as well and want to ensure your business is successful.

I started with Textmate during the Rails craze and transitioned to Sublime due to the portability of themes etc. I think I'm still using my 10 year old TM theme.

Thank you for the continued dedication to this project.

Update: Turns out I bought Sublime in 2012 so I was due to pay the $30 upgrade fee. Gladly! Just completed the process.

Amusingly, your 10 year old TM theme probably shows different/inaccurate colours than it did in TM unless you went to the trouble of converting it to sRGB for Sublime.

I'm out now but it looks identical. I'll share a screenshot and download link when I'm back! It's my favorite.

Here's an interminable github issue thread about it. https://github.com/SublimeTextIssues/Core/issues/880

Update: It's called Made of Code.

Here is the screenshot: https://i.imgur.com/JvDEBk7.png

Here is the theme: http://fl-mwhalen-dev.s3.amazonaws.com/Made%20of%20Code.tmTh...

Please explain your wizardry.

It's just your theme opened in TextMate and Sublime. Sublime displays the AppleRGB colorspace colors (the default in the original TextMate) as sRGB. So they come out off. Check out the linked github thread for the gory/inconclusive details.

This was the easiest upgrade to consider in my career; $30 is nothing compared to the value and ecosystem around Sublime Text.

I'd entertain such a model as OP to ensure the ongoing viability of the app.

I concur with this. I'm appreciative of Atom -- and recommend it to folks who aren't professional programmers and can't imagine paying for a text editor -- and I've heard similar great things about VS Code, but ST and its ecosystem has just been so dependable for me. It was inconvenient to switch over from TextMate to ST initially, but ended up being a great move overall. I haven't yet seen anything on the horizon that would seem like the next phase from ST for me.

I left VIM after 20 year to VS Code. I find that the ecosystem at VS Code to be the best I have ever seen. Great job by Microsoft and I use it on my Linux box all day.

Vim code in VS code works very well (and integrates commonly used vim extension such as 'surround'). You can have the best of both worlds.

I should have stated that I still use VIM mode and bindings. It is the reason why I am there. I tried Sublime, Atom and other text editors and VS Code's VIM mode was on par.

I have tried vscode and vim plugin but found it to be noticeably less responsive than vim.

Is it built in, or is it a plugin?


I like it except that it installed it's own apt sources for updates without telling me, and that rubbed me the wrong way.

I believe that would be distro dependent. On my Linux base I just use the RPM until I added a repo.

Yes. They could make some kind of small cloud service to pay for on a monthly/yearly basis. I would gladly pay for a service that would allow me to share my Sublime settings, addons and projects across all my Sublime installations. This can be done either way but I would happily pay Sublime devs for that.

You should buy multiple licenses of Sublime Text 3.0

I'm in a situation where if I had $10 or $80 extra, I would use it to buy baby supplies, not software. Many others are in the same situation, or worse (thinking of those in developing countries where $80 represent a week's worth of savings).

I agree on the point that useful software/service providers must have a sustainable business model, but I see your point about $80 being too cheap as just one data point (not that you were claiming otherwise, and you did appreciate the value that you get). There are many people for whom spending $80 every few years may be a bigger burden too. Subscriptions, without significantly lower per year prices, also favor the ones for whom these aspects don't matter as much.

I just wanted to put these points across since a single price probably may not well serve all user classes and the developer/company.

I would also like if the developer would agree to release the source code under a FLOSS license should he find working on Sublime Text is no longer worthwhile for him (could be financial but could also just be because he has lost interest in developing it, etc).

Feel free to donate licenses to university students!

I don't feel like Sublime Text has the development momentum to justify the subscription model.

That isn't necessarily bad; it's a great editor and I've been happy to pay for all versions of it.

But if I am paying an ongoing, continual fee, I expect rapid delivery of updates and new features. I pay for an IDE from JetBrains, and they deliver solid incremental improvements and bug fixes, continuously. So I feel the continual fee is acceptable.

Sublime Text is known for going on hiatus for months at a stretch, with no or negligible updates. That would irritate me as a paying subscriber to the software; it doesn't bug me when I pay a lump sum for the big milestone releases.

I think I'm going to start buying licenses for my interns and students. It's an investment in our future. Today's interns and students are tomorrow's senior devs. Best to start them off proper.

Same here, I get wayyyyyy more than $80 of usage out of this software. I'm very very appreciative of the great work that has gone into this product and don't want it going any place!

I think the price is fair given their competition is foss.

Agreed, I just paid my upgrade fee and it seemed too small.

I feel ST could have a checkbox during checkout to add an additional $50 or something in that range. "I'd like to be a supporter for an extra $50". I'm pretty sure I would opt-in to that, but maybe I'm just saying that now because there isn't such an option....

Yes, happy to pay more often. Or just have smaller releases. It must have been years since I last paid for Sublime

You could buy some additional licenses and donate them to someone.

Also, I don't see Sublime in the MacOS App store. I'd suggest it get added there, even if you have to charge more due to the percentage that Apple takes.

I suspect it wouldn't work with the sandboxing required for the mac app store.

The sand boxing requirements plus STs plugins may make that a non starter.

Ah right. Thanks

Honestly, I'm in. I'm absolutely dependent on Sublime Text for my workflow. I don't mind paying more for it /whatsoever/

Would a crowdfunding campaign help to fund a product like Sublime Text which has a lot of traction? I think so.

It has worked in the past, for example: I would consider the recent FontAwesome 5 kickstarter campaign a huge success both from a funding point of view and the backers getting what they wanted. This was only really possible in my opinion due to the reputation Dave Gandy built up by releasing the open source version of FA prior. One needs only to look at the myriad of failed software kickstarters for examples of what happens when a unknown person who hasn't proven they are capable of delivering what people want tries the same thing.

Probably they just need to add a Donate button for people who wants give more than license fee only.

The first thing I did after Ulysses went the way of subscription was get a refund from App Store. :)

Do you think that TextMate fell by the wayside because of an unsustainable business model?

It fell by the wayside because 2.0 was the HL3 of text editor releases, IMO. Open sourcing it was too little, too late. Even though Sublime Text has had long periods of inactivity, the beta releases every quarter or two were just enough to keep people who loved the editor hooked. TextMate lost people like me because there was, for a long time, zero indication it would continue being developed.

Encourage them to set up a Patreon page? Do people use Patreon for software projects?

Or keep the current model as is and add an optional recurring donation somewhere.

You can always buy more licenses and give away to other people. :)

Just buy another license.

Buy a copy for a friend!

nice try sublime dev :>

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