edent a year ago

Having worked for some startups of varying success...

1. Why? What's the benefit for them / their investors?

2. Would a competitor build on top of this and make a success? Why would I want to help them?

3. Does anyone care about this? Most startups aren't solving innovative problems.

4. Who owns the rights to the code? Some of it may not be their IP.

5. What's the liability? Has an API key been left in the code? Could a scammer use the assets to defraud old customers?

6. Perhaps we want to try again in a few years? Let's keep the code on ice.

Don't get me wrong - I'd love to see more organisations do this. But there is very little advantage for them, and a considerable amount of risk. So it is not surprising that they don't.

PaulHoule a year ago

Some of them do.

Years ago I developed a system that would convert the Freebase data dump from a proprietary format into industry standard RDF with complete referential integrity that you could query with SPARQL.

Freebase came out with an inferior official RDF dump that lacked referential integrity and stopped publishing the old data dump. Of course they got bought by Google and shut down in the end.

There were numerous things wrong with my business but in the end when the conversion code had no direct value I open sourced it. It can chew through large data dumps with a map/reduce cluster so somebody might have found some use for it.

senttoschool a year ago

Open sourcing requires a ton of work. You have to rewrite a lot of code that is custom, one-off, hard coded. You'll then have to write a lot of documentation and provide some support to get it off the ground. Lastly, there could be potential legal ramifications as well.

Most startup code relies on 3rd party services as well which doesn't make open sourcing viable.

If you're a dying startup, the last thing you want to do is spend your remaining resources on open sourcing your code.

I don't think open sourcing failed startups is that common. Some do. But the vast majority do not.

BMc2020 a year ago

I had an idea years ago if I was going to create a startup I would call it 'Abandoned Web Site' so we could skip straight to the end.