Thread by @mattklein123

One more thought on @skamille's post. She posits that some of the best projects have come out of non-"pure tech" companies. I think this is correct, and I agree with her reasoning: tight focus, internal customer obsession, and a big enough team to make a bet. 1/

However, one thing that I have noticed is that there seems to be a sweet spot on team size in which this type of creation becomes possible at a non-"pure tech" company. 2/

The org has to be large enough to support risk taking, but not so big that the infra team has started to shift focus towards internal support and has a difficult time keeping their head above water. Related post from me if you haven't read it: 3/

medium.com/@mattklein123/…

I really do believe that many things have to perfectly align to produce an infra tech like Kafka, Envoy, etc. and this type of alignment will become increasingly rare as newer products are built from the ground up on public cloud "serverless" tech. 4/

In the 10-15 year time frame, will the move to public cloud services lead to generally more reliable end-user products and services? I think absolutely yes, so from that perspective I think it's a good thing. 5/

However, we are deluding ourselves if we don't think the big 3 will increasingly drive the infra tech agenda, using OSS when it suits them, proprietary otherwise, and potentially doing away with situations that led to innovations like Kafka and Envoy in the first place. 6/

As is probably clear, I have a lot of thoughts and very mixed feelings about this situation. 😀 fin/