To be honest, the repositioning isn't really a response to the brain drain -- it's a reflection of a changing culture at Basho that was the cause of the brain drain in the first place.
The primary reason I left Basho is that I felt the company was turning into a "don't invent here" culture, the polar opposite of "not invented here". The goal was no longer to solve hard distributed systems problems and build amazing technology, but to just integrate various "trendy" technologies and make an enterprise simplification play.
The problem is that most of these other technologies have major failings and/or punt on the corner cases that us old time Basho engineers obsessed over.
The whole point of Riak was to be the most highly available, fault tolerant, trusted database you could use. You can't just integrate Riak with arbitrary products X, Y, and Z without compromising on those core tenants.
I really hope Basho can deliver on the promises they're making about this data platform product. I really do. But, for various reasons I can't really talk about, I'd say I'm extremely skeptical.
(BTW, for those that don't know me: I'm a former Principal Engineer at Basho and was the lead developer on a variety of sub-systems in Riak over the last four years)
Thanks for the context - I'm not sure if we've met, but I definitely know and respect your work. Of course there are compromises with this approach, but unfortunately I need a set of capabilities that are not currently available from a single source. Ultimately this just another entrant in a field of options, some of which have evolved in a similarly ad hoc fashion (though one I'm looking at has been under development for 25 years). I hope it can at least hold its own.
Well they have to monitze effectively if they expect to retain talent, and so it's definitely related in that way.
It's a nice database, but not well understood. I've found it difficult to communicate how everything fits together to people do don't use this stuff everyday, so maybe this will help? At least it makes it clear that a seemingly complex configuration is perfectly valid and workable in practice.
jtuple|10 years ago
The primary reason I left Basho is that I felt the company was turning into a "don't invent here" culture, the polar opposite of "not invented here". The goal was no longer to solve hard distributed systems problems and build amazing technology, but to just integrate various "trendy" technologies and make an enterprise simplification play.
The problem is that most of these other technologies have major failings and/or punt on the corner cases that us old time Basho engineers obsessed over.
The whole point of Riak was to be the most highly available, fault tolerant, trusted database you could use. You can't just integrate Riak with arbitrary products X, Y, and Z without compromising on those core tenants.
I really hope Basho can deliver on the promises they're making about this data platform product. I really do. But, for various reasons I can't really talk about, I'd say I'm extremely skeptical.
(BTW, for those that don't know me: I'm a former Principal Engineer at Basho and was the lead developer on a variety of sub-systems in Riak over the last four years)
rch|10 years ago
deegles|10 years ago
wlesieutre|10 years ago
I can't speculate on how that all relates to brain drain.
InvisibleCities|10 years ago
rch|10 years ago
It's a nice database, but not well understood. I've found it difficult to communicate how everything fits together to people do don't use this stuff everyday, so maybe this will help? At least it makes it clear that a seemingly complex configuration is perfectly valid and workable in practice.