About Us

Human written

How hard can it be to build a forge platform that scales

^ No idea

Motivation to make a serious effort at a forge platform came from dislike and distrust of: Github, Microsoft, and USA.

I do think Github is a great product, and it is very unlikely I will stop using it for FOSS projects. And at their scale, I can’t even imagine the operational costs. Kudos to the engineers at Github!

But nothing is perfect, and we try to deliver an excellent forge platform. No double platform costs, and something we can say will scale with

Distributed Git seemed a really fun problem to solve, and over the year we have learned more than we thought about how to do this. The initial version of neosource was putting every single operation/commit in foundationdb, which seemed smart at the time. But over the time we simplified and understood the inner workings of Git, and made something novel and performant.

Other forges have been showing up since we decided on our architecture, and each time we were eager to read their technical documentation about their systems. Only to realize everyone came to the somewhat same solution, albeit incomplete or worse.