Interesting post - I'm coming from a COSS database perspective (TerminusDB), so talking my book. The point of COSS imho is to create broad value and monetize a

Interesting post - I'm coming from a COSS database perspective (TerminusDB), so talking my book. The point of COSS imho is to create broad value and monetize a percentage of that value. A few points: 1. In database land, there is no other available venture scale business model. If you look at the top 10 databases on DB-Engines (https://db-engines.com/en/ranking https://db-engines.com/en/ranking), the most recent proprietary db is SQL Server which launched in 1989. Yes, databases are sticky (we see you Oracle), but unless you believe there are no further investable plays in databases, then open source is the only way to invest. (Snowflake is an exception to the rule, but in this case, it is just that). 2. Useful recognition that offering hosted versions of the product is a viable business model that can deliver SaaS type margins. Mongo's Atlas is the obvious leader here - a huge and fast-growing revenue stream. I would also mention the companies that have built successful SaaS/marketplace offering on OSS. Informative list of all the COSS companies above 100m in revenue: https://docs.google.com/spreadsheets/d/17nKMpi_Dh5slCqzLSFBoWMxNvWiwt2R-t4e_l7LPLhU/edit#gid=0 https://docs.google.com/spreadsheets/d/17nKMpi_Dh5slCqzLSFBo... 3. I find this conclusion difficult to swallow: 'The ratio of failed OS businesses to successful ones is worse than in prop-tech; revenue kicks in much later'. I can't find any evidence for the first point - it needs substantiation. For revenue, lots of the most successful venture backed businesses (facebook) have revenue kicking in later so they can maximise growth potential – that’s seems ideal for venture!