> This IMO really highlights the power of having an ecosystem built around tools like tree-sitter because they allow for powerful dev UX tools to be more democratized
LSP is also an example of this and it is also an example of how dangerous it can be when these tools are backed by big companies: https://news.ycombinator.com/item?id=31760684
If microsoft, tomorrow, pulled the plug on the pylance LSP and removed all access, and the only thing left is the opensource version, the world is still at a better position than if they hadn't invested any money or opensourced anything.
This is strictly different from google owning chrome, because with that ownership, they have the ability to dictate protocols for the web.
The LSP is a protocol, which now is quite widespread, and will continue to survive and be enhanced, regardless of microsoft's involvement or not. The fact that they can choose to make some language servers close-sourced sucks, but being able to access a closed source component is _still_ strictly better than having zero access in the first place.
Microsoft has the resources and ability to make better language servers which could be closed source now, so people end up switching to those tools, owned by Microsoft, with better language servers, killing other editors if they are not able to catch up.
E.g.: Why would a C# programmer use Neovim if the language server is worse than the one that VSCode has? (Which is now proprietary and closed-source). The difference right now is insignificant, but the future tools and features that they are adding to the new proprietary C# language server will not be available for other editors.
[1] https://en.wikipedia.org/wiki/Embrace,_extend,_and_extinguis...
they can only extinguish by forcefully bankrupting alternatives. Unfortunately for microsoft, there's no such thing as bankrupting an opensource project.
If neovim's feature set can't keep up with the might of microsoft's development budget for vscode, and many users switch over, then it's a net-win for those users. If microsoft then removes the C# LSP afterwards (or charges money for it), neovim is still around. Those users who don't want to pay can move back to neovim (and may be someone would then start contributing to it again, and over time, the features of neovim will improve again).
Microsoft killed netscape by depriving it of a business model. But you cannot deprive opensource of a business model, because that model is not profit oriented.
Unless microsoft patents the LSP protocols and stops other people from using it (and even then, other editors do not die, unlike netscape), there's no real way for microsoft to EEE.
Google can deprive firefox of a business model (directly by stopping their payment to mozilla, and indirectly, by making the web protocols so restrictive that firefox cannot implement them properly, such as DRM). Therefore, google is in a position to EEE other browsers.
Lots of open source does, in fact, come out of a profit oriented business model. Open source is harder, but not impossible, to kill by killing the business model of whoever runs the current project because open source provides rights to other people to pick it up even if the original sponsor’s interest becomes to end it, so as long as there is someone or some group with the resources and interest, it can continue.