Everyone shitting on D3 here should stop and consider that there must be some fundamental reason why a library like D3.js has withstood the test of time in the JS landscape, when most libraries go out of fashion within months.
Most of the comments here complaining about D3 boil down to the following:
- Using the wrong tool for the job... if all you want is off-the-shelf visualizations, use one of the many visualization libraries out there that utilize d3.js on the back-end
- Not wanting to put in the effort to learn the basics of SVG, CSS and the D3 API, and yet expecting hand-holding and everything to just be easy
The beauty of D3 is how powerful it is and how unopinionated it is. You can literally make any type of visualization with it, and it has some astoundingly useful helper libraries for data binding, data manipulation, color handling, etc.
As someone who picked up D3.js from scratch, I learned it by having clear goals regarding what I wanted to achieve. Having built charts / visualizations that are used in several products, it became very clear, very early on, that unless your customer/product requirements are super light-weight and run-of-the-mill visualizations are okay, you are bound to get to a point sooner or later where a visualization library isn't going to cut it. At that point, you're either going to spend a ton of time trying to shoe-horn a feature into an existing library that isn't made to be flexible, or you could have just saved yourself a whole bunch of effort by having gone the route of writing your visualizations in D3 from the get-go.
D3 absolutely has a place in the data visualization space. You just need to be aware that it may not be the right tool for you, and if you do need to wield it, you need to put in the effort to actually familiarize yourself with the D3 related landscape rather than expecting to be able to simply copy-paste examples and hack your way to a working solution.
PS - I should mention, that I'm an ML Engineer / Data Scientist, not a data viz geek, so it's not like D3 is relegated to people who do data viz for a living. There are plenty of folks who have picked up D3 on the side and done amazing things with it.
Not wanting to put in the effort to learn the basics of SVG, CSS and the D3 API, and yet expecting hand-holding and everything to just be easy
Sadly, this problem has become endemic within the web development community in recent years. People are more concerned with learning the latest buzzword framework than the fundamentals.
The article itself quotes this tweet early on:
The trouble with D3 is to build a visualization you must also have a deep understanding of SVG, DOM, JavaScript, geometry, color spaces, data structures, the standard model, and quantum physics — @seecmb
Well, yes. Designing and implementing good visualisations is a skill. It has many aspects, and they aren't trivial. If you want to draw a standard, no-frills, minimally-customised chart then there are many higher-level libraries that will draw one for you and you're not the target audience for D3. If you want to have more customised and creative visualisations and interactions, you have to customise and create. That requires understanding the tools available and knowing what you're trying to achieve with them, but if you are willing to put the work in to reach that point, D3 offers a useful and powerful toolbox.
Most of the comments here complaining about D3 boil down to the following:
- Using the wrong tool for the job... if all you want is off-the-shelf visualizations, use one of the many visualization libraries out there that utilize d3.js on the back-end
- Not wanting to put in the effort to learn the basics of SVG, CSS and the D3 API, and yet expecting hand-holding and everything to just be easy
The beauty of D3 is how powerful it is and how unopinionated it is. You can literally make any type of visualization with it, and it has some astoundingly useful helper libraries for data binding, data manipulation, color handling, etc.
As someone who picked up D3.js from scratch, I learned it by having clear goals regarding what I wanted to achieve. Having built charts / visualizations that are used in several products, it became very clear, very early on, that unless your customer/product requirements are super light-weight and run-of-the-mill visualizations are okay, you are bound to get to a point sooner or later where a visualization library isn't going to cut it. At that point, you're either going to spend a ton of time trying to shoe-horn a feature into an existing library that isn't made to be flexible, or you could have just saved yourself a whole bunch of effort by having gone the route of writing your visualizations in D3 from the get-go.
D3 absolutely has a place in the data visualization space. You just need to be aware that it may not be the right tool for you, and if you do need to wield it, you need to put in the effort to actually familiarize yourself with the D3 related landscape rather than expecting to be able to simply copy-paste examples and hack your way to a working solution.
PS - I should mention, that I'm an ML Engineer / Data Scientist, not a data viz geek, so it's not like D3 is relegated to people who do data viz for a living. There are plenty of folks who have picked up D3 on the side and done amazing things with it.