| ▲ | bsder a day ago |
| > MS is under no obligation to provide an OSS editor, but playing tricks after luring people in is not nice. Microsoft is partly to blame, but people have been warning about this over and over and over ad nauseam and people still choose to use VSCode. You couldn't even get people to not use the proprietary extensions for C/C++, Python and remote development. The problem is that Microsoft dedicates enough resource to development that everybody else looks like a rounding error. For example, anybody could have produced the Language Server Protocol, but nobody had the critical mass until Microsoft shoved it down everybody's throats. Until somebody puts a significant amount of money behind an alternative, Microsoft is going to continue to win this battle. (I was going to also say "or the OSS guys all unify behind a choice" but Hell will freeze over before that happens.) |
|
| ▲ | harshitaneja 14 hours ago | parent | next [-] |
| Genuinely curious, how did Microsoft "shoved it down everybody's throats"? And weren't Jetbrains, Eclipse, Vim, Emacs dominant enough(especially Jetbrains) to have done so before Microsoft? |
| |
| ▲ | bsder 12 hours ago | parent [-] | | VSCode garnered such a significant market share that supporting VSCode almost immediately becomes your #1 editor (see Rust surveys: 2016-3%, 2017-30%, 2018-45%, 2020-54%, 2023-62%--note: VSCode was only released in 2016!). Since the LSP was the only effective way that you could support syntax highlighting in VSCode, languages had to create an LSP or they didn't exist to VSCode users. Once the language supports VSCode, anyone not already steeped in the editor wars switched. At that point, the editors had to support the LSP or get left behind. With JetBrains, the issue is that they would have almost certainly considered an LSP as a competitive advantage. It would have taken some amazing foresight to release something like that as OSS and not be afraid of losing market share if you get vim/neovim to adopt it (you can ignore emacs market share--the editor wars are all but over and emacs lost badly). | | |
| ▲ | mdaniel 4 hours ago | parent [-] | | > With JetBrains, the issue is that they would have almost certainly considered an LSP as a competitive advantage As a point of comparison, the value that I get from JetBrains products is about 35% UX niceties[1] and 1500000% code intelligence, which no LSP that I have ever seen even strives to do that. They all appear to be focused on "jump to declaration", "what can I type here", and some of them ferry "linter" results back to the user but that is just using LSP as a conduit not that the LSP itself is doing anything intelligent That's why I throw up in my mouth when someone claims their vim+lsp is a python powerhouse because, sure, it's better than nothing but even PyCharm open source blows the doors off of any LSP 1: and even that has been under constant attack over the past few years from their gravely misguided product management team, culminating in the "you are committing code wrong" https://news.ycombinator.com/item?id=43710699 |
|
|
|
| ▲ | owebmaster 21 hours ago | parent | prev [-] |
| > (I was going to also say "or the OSS guys all unify behind a choice" but Hell will freeze over before that happens.) The editor war is going as strong as ever, emacs vs vim will still be here in 20 years. Compared to 10 years ago, the amount of people using emacs and vim only grew, although VSCode growth was 1000x faster. |
| |
| ▲ | skydhash 17 hours ago | parent [-] | | I was watching the "Simple Made Easy" talk by Rick Hickey, and while he was talking more about programming languages, the talk could extend easily to editors and other type of tooling. People are always going for easy, not simple. Vim is very simple (a composable language for editing, straightforward integration with cli tools, easily extendable,...). Emacs is simple (Major mode that dictates main operations and display, minor modes for additional features, integrations between them can be described more as a complex web than a simple graph...). VS Code is easy (helpful suggestion for plugins, Familiar IDE-like interface, default setup, ready to hack on projects,...), but scratch that surface and the complexity appears (behemoth web engine, settings all over the place, app store like marketplace, extension are full blown software project,...). All the cons of IDE with none of the pros. | | |
| ▲ | throwy63658 15 hours ago | parent | next [-] | | People will opt for what's accessible. Vim requires you to learn an alien control scheme to even navigate through code, and Emacs requires you to learn an incredibly unique programming language — one that sees no real world use, let's be honest — for basic tasks. Is it any surprise, then, that Microsoft had no issue swooping in and dominating the editor market when Emacs and Vim users believe that these are "features"? VSCode does a great job keeping the complexity away from the user. Vim and Emacs shove it in your face. I prefer VSCode's approach. | | |
| ▲ | teddyh 2 hours ago | parent | next [-] | | > Emacs requires you to learn an incredibly unique programming language — one that sees no real world use, let's be honest — for basic tasks. Exactly what basic tasks in Emacs requires Emacs Lisp? Give me your best/worst example. I can’t think of one. | |
| ▲ | genewitch 9 hours ago | parent | prev [-] | | I am not proud to admit that basically the only thing I know how to do in vim is edit the network configuration so I can install nano - in much the same vein as using iexplore.exs to download firefox back in the day. |
| |
| ▲ | bsder 13 hours ago | parent | prev [-] | | "Not broken" trumps both simple and easy. Syntax highlighting on the open source editors was a pile of regex fail before VSCode came along and forced everybody into the LSP. |
|
|
