Author : Yong Cui

October 6 | 6 min

Although major python IDEs (e.g., Visual Studio Code, PyCharm, and recently DataSpell) all support editing Jupyter Notebooks, JupyterLab is still my go-to IDE when I use notebooks for my data science projects. As an interactive IDE, JupyterLab allows us to visualize the data (e.g., tabular data, images) step by step in a cell-based organization. When we use JupyterLab, there are several ways to launch it — either using the terminal or using the Anaconda Navigator app. In either case, you’ll have to launch another app before you can run JupyterLab. Although it’s not a big deal, the extra step represents inconvenience to some extent. To solve this issue, the JupyterLab team has been working to make it better. About less than one week ago, in Jupyter’s blog, they made an announcement on the release of the JupyterLab’s desktop app. The link to the blog article is provided below for interested users. It should be noted that the download link isn’t available on the Jupyter’s web page. Instead, it’s available on the GitHub page. The JupyterLab development team made a smart choice by building the desktop app using electron as electron allows programmers to develop cross-platform desktop apps with web development languages, including JavaScript, HTML, and CSS. Using the electron as the tool for building the front-end, JupyterLab Desktop App is thus a cross-platform independent IDE. The installers for major operation systems all are available on the GitHub page. Windows: Windows Installer Mac: macOS Installer Linux: Debian, Ubuntu Linux Installer & Red Hat, Fedora, SUSE Linux Installer To give it a try, I downloaded the macOS installer on my MacBook laptop. The installation is painless, and you just need to follow the prompts and install the app in the desired destination.

As you can see, it has the same user interface as what you can see on a web browser. So if you’re familiar with JupyterLab web interface, it takes no time for you to start your notebook editing with the same level of user experiences. In terms of the app’s menu, it just provides the bare minimal commands that any Mac app has to offer, and thus, there is a great room for the developer team to add functionalities to the app. Certainly, it also requires extra work for them to accomplish this, because JupyterLab Desktop App is just like a UI wrapper to serve as a container for JupyterLab. Thus, I bet that they’ll be still focused on the core JupyterLab development, and I doubt they’ll do much improving the desktop app beyond what it has to offer now.

Final Thoughts

Anyhow, for experienced JupyterLab users, it won’t be a too much game changer. What it can do is probably saving you a step by allowing you to launch JupyterLab directly. However, for JupyterLab beginners or data science beginners, I doubt it will be very attractive, because they’ll have to install the desktop app and conda separately to make it to work. Why don’t they simply go install Anaconda? It will take care of the conda environment, and Anaconda Navigator offers a shortcut for you to install and run JupyterLab very conveniently. Thanks for reading this article. Stay connected by signing up my newsletter. Not a Medium member yet? Support my writing by using my membership link.