With Iniobong Benson and Hamish Cambell

Listen

Show Notes

Kart provides distributed version control for geospatial and tabular data. It  stores geospatial and tabular data in Git, providing version control at the row and cell level, it also provides repository working copies as GIS databases and files, and allows edit to be done directly in common GIS software without plugins.

Features of Kart

• Kart stores geospatial and tabular data in Git, providing version control at the row and cell level.
• Uses standard Git repositories and Git-like CLI commands. If you know Git, you’ll feel right at home with Kart.
• Provides repository working copies as GIS databases and files. Edit directly in common GIS software without plugins

What is  Kart?

Kart is an open-source software project. That means it is freely usable everyone can look at the source code and make modifications and use it without any. It is an open-source software project for data version control for geospatial data. The purpose of Kart is to make it easy to store your spatial data in a way that you can make changes to it, and see those changes and push and pull those changes to other people so they can have it on their data as well. So we can also see, the way that people work in the GIS world, we record a lot of files on your disk or the one you could share and we have this sort of problem of how you manage all the copies of your data and kart is out to solve that problem.

Tell us a little about Kart’s history - it was formerly called Sno, why was the name changed?

Yes, there’s an old joke in software development, - there are only two hard things in computer science; session validation and naming things, and its sort of the naming things that we are talking about. We have a convention internally of using names of flowers for working names for projects. When we start up a new thing and we need a name, so we’ll put the name of a flower, and Kart actually started off as “Snowdrop”, but as software developers, we’re also lazy with typing so we just said Sno -three letters -  and we really liked that name and got used to it and could type it really quickly. But unfortunately, we found that users were confusing that with some Snowflake DB, which is a new-ish but very popular database solution out there at the moment, so yeah we had to go and find a new name which was a difficult problem. After a lot of discussions and arguments, we settled on Kart, that's K-A-R-T. And that has been good. It has an association with mapping and the word cartography, but most importantly it was quick to type, oh yes, it's stuck now and we’re used to it, yeah it's a good name I think

The purpose of Kart is to make it easy to store your spatial data in a way that you can make changes to it, and see those changes and push and pull those changes to other people so they can have it on their data as well

Distributed version control in software development, what is that? And how does it relate to geospatial data?

Version control is the ability to keep different copies of your data or code or whatever you are working with together in one and get to the different pieces and manage the changes over time. The reason we got into it is because at Koordinates, we primarily help our customers publish lots of data and we help end users consume lots of data… But it's a one-way data flow and changes hardly ever get the other way. So if you are working on changes to the data you want to publish, managing the changes you’re making to that, and preparing releases is still quite a difficult problem for the end user too. There are some tools out there, but they can be very expensive and they don’t work particularly well for particular projects that span lots of people.