When comparing Visual Studio Code vs Jupyter, the Slant community recommends Visual Studio Code for most people. In the question“What are the best Python IDEs or editors?” Visual Studio Code is ranked 1st while Jupyter is ranked 8th. The most important reason people chose Visual Studio Code is:
Visual Studio Code comes fairly complete out of the box, but there are many plug-ins available to extend its functionality.
Specs
Ranked in these QuestionsQuestion Ranking
Pros
Pro Extendable through plug-ins
Visual Studio Code comes fairly complete out of the box, but there are many plug-ins available to extend its functionality.
Pro TypeScript integration
There is very solid TypeScript integration in Visual Studio Code. Both are developed by Microsoft and VSC itself is written in TypeScript.
Pro Integrated debugging
VSC includes debugging tools for Node.js, TypeScript, and JavaScript.
Pro Ready to use out of the box
You don't need to configure and add plugins before being productive. However, you can add plugins if needed but for the basics you're well covered.
Pro Integrated terminal
There's no need to press alt+tab to go to a terminal: it is directly integrated into the editor. Shift+~
is a handy hotkey to toggle the integrated terminal.
Pro Great performance
For a 'wrapped' web-based application, Visual Studio Code performs very well.
Pro Libre/open source
Released under the MIT License.
Pro Fast and powerful
VS-Code has the speed of Sublime and the power of WebStorm. Perhaps this is the best software that Microsoft has ever created.
Pro JavaScript IntelliSense support
JavaScript IntelliSense allows Visual Studio Code to provide you with useful hints and auto-completion features while you code.
Pro Embedded Git control
Visual Studio Code has integrated Git control, guaranteeing speed, data integrity, and support for distributed, non-linear workflows.
Pro Updated frequently
There's a new release of Visual Studio Code every month. If you are one of the insiders then releases are daily.
Pro ESLint integration
ESLint integrates great. You can define your rules trough .eslintrc.* as usual and vs code will autofix your code on save. So your code is always in style.
Pro Extensions (aka plugins) are written in JavaScript
Extensions are written in either Typescript or JavaScript.
Pro Active development
It's really nice to see how the code editor evolves. Every month there is a new version with great communication of new features and changes.
Pro Integrated task runners
Task runners display lists of available tasks and performing these tasks is as simple as a click of the mouse.
Pro It has gotten really good
All it takes is one stop for all the features many people need.
Pro Custom snippets support
Snippets are templates that will insert text for you and adapt it to their context, and in VSC they are highly customizable.
Pro Huge community behind it
The ease of getting assistance and finding tutorials is increasing as the community grows.
Pro JS typechecking
It leverages TypeScript compiler functionality to statically type check JS (type inference, JSDoc types) with "javascript.implicitProjectConfig.checkJs": true
option.
Pro Python support
Excellent Python plugin, originally created by Don Jayamanne, now hired by Microsoft to extend and maintain the extension.
Pro Good support for new Emmet syntax
Pro High fidelity C# plugin
The Omnisharp plugin is very powerful providing full sln, csproj, and project.json support.
Pro Support RTL languages
It supports pretty web rtl languages like arabic languages when most of other editors don't support it.
Pro Inline definition picking and usages finding
These features allow you to have a glance at code without opening it as a whole in a separate tab. Moreover, editing is allowed.
Pro Web-based development allows for usage literally anywhere
Because the editor is a web app (the Jupyter Notebook program is a web server that you run on the host machine), it is possible to use this on quite literally any machine. Morever, you can have Jupyter Notebook run on one machine (like a VM that you have provisioned in the cloud) and access the web page / do your editing from a different machine (like a Chromebook).
Pro Interactive
Most IDEs require you to separately run Python to see the output of a particular piece of code. By contrast, Jupyter Notebook can evaluate Python statements inline, giving you the immediate feedback of interactive use of the interpreter while keeping your changes saved.
Pro Graphing , charting, and other math/numeric capabilities
The interactive editor is able to display complex equations, charts, graphs, etc. making this particular editor very well-regarded among data scientists.
Pro Open source
Because it is open source, you can review the source code and also propose extensions and fixes to it. It is also possible to fork the repository and make changes to it to customize it for your specific use case.
Pro Supports multiple different programming languages
Jupyter Notebook, formerly known as ipython, used to be specific to Python; however, in recent iterations, it has become capable of general purpose usage for any programming language. Thus it is possible to use this and have a consistent developer workflow, regardless of language.
Cons
Con Embedded Git isn't powerful enough
You can do nothing but to track changes, stage them and commit. No history, visualization, rebasing or cherry-picking – these things are left to git console or external git client.
Con The autocomplete and code check is not as powerful as the one on WebStorm
Sometimes it doesn't tell you if you made a typo in a method name or if a method is not used and several other important features.
Con File search is extremely slow
It's absolutely not possible to use this tool with big projects given how long it takes to search for files.
Con Project search limits results
Because file search is so slow your results are limited in order to simulate a faster search.
Con Very bad auto import
Con Generalized
VS Code is a general code/scripting IDE built to be lightweight and for people familiar with their language of choice, not directly comparable to Visual Studio in power or scope.
Con Memory hog
Allegedly, VS Code is "lightweight". Yet, running multiple instances of it at once, you may get many "out of memory" messages from Windows despite 16 GB RAM. (While of course also running other things. The point is the comparison with some other IDEs/editors where running them alongside the same number of other applications doesn't cause Windows to run out of memory)
Con Poor error fix suggestions
Error detection and suggestions/fixes are poor compared to IntelliJ platforms
Con A "me too" offering from MS, far behind other well established editors that it attempts to clone
Other IDEs specific to a language often offer better tools for deep programming.
Con Slow launch time
Slower than it's competitors, e.g. Sublime Text.
Con Emmet plugin often fails on even simple p tags
Con Have no good default js style analyzer
In WebStorm there is analyzer that checks for warnings and highlight this in yellow, here you cannot find or add it even with plugins. It is possible to have it as errors with linter but while you are actively changing file that's not very nice.
Con .sass linting is terrible
Con Is not an IDE, is a text editor
Con Notebook-style makes reusing functions annoying
Con Interactive usage takes some getting used to
While the interactiveness is extremely, extremely powerful and useful, it does take a little bit of work getting to a point where it is "normal".
Con First time setup is more difficult than for other IDEs
Since Jupyter Notebook really requires two programs (the server and your browser) getting things setup in a way that works for you is a little more complex than for an ordinary IDE. For example, if you run the server and edit on the same machine, creating a little wrapper script that starts the server and then launches the browser pointing to it and gives an icon to this script is a small amount of setup but is more involved than a simple installer for other IDEs. Likewise, if you do remote development, creating a URL that will lazily spawn the Jupyter Notebook server and then turn it down when it is no longer in use is also a little bit of work to setup.
Con Non-trivial security configuration for remote access
By default, the editor is only accessible from localhost; however, if you want to run Jupyter on a VM in the cloud and do your editing through a web browser on a different computer (e.g. a Chromebook), there is some non-trivial security work to ensure that it is set up in a secure manner.
