I miss OriginPro in my undergrad when we had campus licenses for, before moving to matplotlib for data visualization. matplotlib is simply too disappointing for making publication quality figures. The most recently encountered problem is how to plot with a broken x-axis, which is one of the most basic need in physical science but requires a non-trivial amount of hacking to get with matplotlib.
Open source tool or not, I don't care at all as I get the science right. I have already enough frustration dealing with my samples, so I simply want the least frustration from the software I use to plot.
Matplotlib is a bit painful. Often seaborn will work quicker, especially when using Pandas dataframes with proper column names and seaborn compatible layout.
Its annoying that you cannot create a broken axis out-of-the box, but I am sure you can wrap this to make your own convenience function: https://matplotlib.org/stable/gallery/subplots_axes_and_figu...
That link was what I referred to after Googling, but in my case I need the width of the left part and the right part to be different, which requires setting width_ratios in the subplots and adjusting the slope of the hacky lines used to draw the broken axis symbol. seaborn also would not help in this exact case.
There is a package by some nice guy: https://github.com/bendichter/brokenaxes just to do the broken axis. But not being built-in in Anaconda is already an annoyance, and in my case it generates a figure with a ugly x-label.
I ended up letting ChatGPT generate the code for me with the two required hacks. I simply need the figure in the minimal amount of time and with the least mental bandwidth, so I can focus on the science and catch the conference deadline. Origin is a very "over-engineered" piece of software, but hey getting a broken axis is so simple (https://www.originlab.com/doc/Origin-Help/AxesRef-Breaks ). Sometimes the "over-engineering" is necessary to minimize users' pain.
Honestly, if you're doing scientific work there is no reason not to output the data somewhere and plot in R with the standard lib (insanely good for science style plotting but hard to use) or ggplot (what matplotlib wished it was)
Honestly, when it comes to hacking things together with matplotlib I outsource all of my thinking to chatgpt to do the 80% of doc hunting that is honestly not worth it since everything in matplotlib is labelled inconsistently.