2024 W16 - What should we open-source next?

Humberto

Every week I post about one thing that happened at Rows. We're building in public!

Recently we open-sourced our RowsX extension for Chrome.

That project is hosted on Github so that anyone can check it out and contribute. It's on a MIT license, which is very permissive. We also setup a bounty program to reward contributions, created a page (that same github page) and Bam! It's out now.

Our team is happy with the decision. So, what’s the next thing we should open source?

We have had a few discussions about different ideas:

Translations, so that any user can contribute to the most requested langs (Spanish, Chinese, Portuguese) or really just any lang!
Spreadsheet Functions and Integrations;
A grid component for rendering tables;
Charts, either as a component, plug-in system for new charts, or just for improvements;
The whole platform (!?).

It's a great way to see the pros and cons of being open. We haven't taken any decision. For now, I think we will keep our current plans and bide our time.

RowsX is a tool for business people that grabs data from webpages and loads it into a spreadsheet. At the moment we are not aiming to build a fully automated scraper platform that operates beyond the actions of the user and what the user can see.

That's for a few reasons:

A simple extension like ours gets a job done for business people. You see a table on a page, you click a button, you get the data in a spreadsheet. When you get into the scrapper customization game, you have to get the users to explain which data they want, where it is, how much they want of it, etc. It's much more complex, in fact, it's a whole industry.
It's also not our goal. We have Integrations, and that's the best way to import data. The vast majority of data providers will benefit from having a great experience for loading data into a spreadsheet. And that's what you get in Rows, a direct connection to the source.
It's not very nice. Creating an automated scrapper is a whack-a-mole game with publishers (website owners), as they will always try and block anything that smells of automation. Is the scrolling too smooth? The clicks too timely? are you injecting code on the webpage? Blocked.

AGAIN, what’s the next thing we should open source?

See you next week!

Humberto