# Show and tell: Build beautiful maps in no time with MapFast

**URL:** <https://community.plotly.com/t/show-and-tell-build-beautiful-maps-in-no-time-with-mapfast/88244>\
**Category:** Dash Python\
**Tags:** show-and-tell\
**Created:** [October 25, 2024, 2:27pm UTC](https://community.plotly.com/t/show-and-tell-build-beautiful-maps-in-no-time-with-mapfast/88244 "2024-10-25T14:27:09Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![spriteware](https://sea2.discourse-cdn.com/flex024/user_avatar/community.plotly.com/spriteware/32/29140_2.png) [@spriteware](https://community.plotly.com/u/spriteware)\
**Post date:** [October 25, 2024, 2:27pm UTC](https://community.plotly.com/t/show-and-tell-build-beautiful-maps-in-no-time-with-mapfast/88244/1 "2024-10-25T14:27:09Z")

</div>

Dear community,

In June 2024 I wanted to build a side project. I came up with an idea to solve my own problem : **making customized choropleth maps from data is a struggle**. I wanted to have a tiny software that help me.

So I built [MapFast](https://mapfast.co), a dash app that does not look like the typical dash app: [https://mapfast.co](https://mapfast.co)

 ![3d0402d6_3455_4b7c_823c_474fcc625295_image](https://us1.discourse-cdn.com/flex024/uploads/plot/original/3X/6/8/68fd790ac61fd8cc645ef3d7666d1c1ff9bc66cf.png)  
 ![cfabedf9_5f79_4b5d_980b_6f391689cfbc_image](https://us1.discourse-cdn.com/flex024/uploads/plot/original/3X/0/c/0c7eb48664695a483b4ee514ed772d0f4173a108.jpeg)

## WHY

If you create a choropleth map with Python or R, you’ll need to

1. 🙂 Find the geometries / boundaries to plot. This can take some time for specific regions.
2. 🥱 Then you’ll need to match the geometries with your own data.
3. 😭 After this, you often get a map that is somehow really hard to modify with code : changing labels position, adding a customized title, with customized colors… Good luck.

On the other side, professional software like QGIS or ArcGIS is too complex for simple choropleth maps.  
There must be a “between”. That’s why I built MapFast.

## UNDER THE HOOD

Building this app gave me the opportunity to try out [dash-mantine-components](https://www.dash-mantine-components.com/) instead of [dash-boostrap-components](https://dash-bootstrap-components.opensource.faculty.ai/). I also used a lot of **clientside callbacks** (about 50% of my code is JavaScript for the map interaction). These clientside callbacks interact with [Fabric.JS](http://fabricjs.com/), that enable SVG manipulation and canvas drawings.

## OUTCOME

I learned a lot in the process. It was really satisfying to see that I could do a nice looking app with Dash.  
**It strengthens my belief that Dash can be used for full SaaS.** (Which I also experienced on my main company).

I also learned how complicated it can be to have all geometries simplified properly, that there is no universal dataset of geometries for all regions and all levels. I realized that not adding islands or small territories can really hurt the feeling of some people. I discovered the need for point of view maps (i.e. for debated territories).

As a side project, the app is still buggy. A lot…  
But it works, and I could make maps that I am proud of. Below are some examples.

## EXAMPLES

 ![11_bxl_rent_prices](https://us1.discourse-cdn.com/flex024/uploads/plot/original/3X/9/4/9430156e056431d74b7fbb845ec04eeff07a5c3b.jpeg)  
 ![14_europe_beneficiaries](https://us1.discourse-cdn.com/flex024/uploads/plot/original/3X/8/e/8eefd40a2dd5d4fd8847f5e44793c1879795c166.png)  
 ![6_italy_wine_production_2023](https://us1.discourse-cdn.com/flex024/uploads/plot/original/3X/9/4/947561d262496c652677822db0cb33622bdda792.jpeg)

## WHAT NEXT ?

**I am actually struggling to get feedback on this. I don’t know if it can be useful and for who.**

I suspect that it is still too complicated to use and/or people just want their map without the need for customization.

Some raw ideas I have:

- add different point of views
- embbedable dynamic maps
- automated normalization (e.g. per capita)
- other type of maps (e.g. point map, bubble maps).

* * *

🙏 It would be amazing if you could try it out and tell me how was the experience.

Direct link to the app : [https://app.mapfast.co](https://app.mapfast.co)  
Thank you,

François