Comment by andrewstuart2
7 years ago
As a Red Hat employee (opinions my own, not Red Hat's, etc) and though not involved in any discussions like this, I can see some potential for it being a good move. There's a lot of work being done to a) move things from mainframe to distributed and b) help people squeeze the last bits of utility out of their mainframes. That requires knowledge on both sides of the house.
The world is going to open-source distributed systems built on commodity hardware, which Red Hat has done a great job of building a business model around. For old established companies, though, the migration is a lot of work, and there are _definitely_ still plenty of large companies who have purchased/leased mainframes for long periods of time, and would like to modernize, but also can't afford to throw everything away and rebuild from scratch. There's a lot of work being done already between the two companies to run Go, Docker, Kubernetes, etc, on mainframe, and for companies with mainframe resources, being able to get a little more utility out of those sunk costs is very attractive, and something Red Hat's expertise has (and would continue to, presumably) help accomplish.
That being said, I'm pretty surprised at the news, and I'll be watching closely to see how things go.
This is the most positive comment I've seen in this thread. To the extent that the ensuing development effort finds bugs in container orchestration systems, creates new features, creates new abilities to manage services on "hybrid" private-cloud-plus-mainframe systems, helps to modernize old-school businesses that run infrastructure the global economy relies upon... this could actually be a very good thing for open source and future investment therein. Skepticism is healthy and warranted, but this could be a good match.
It's also worth pointing out that Red Hat (and Big Banks) have been desperately trying to train new mainframe operators with limited success due to the shift to distributed, to the extent that they've even been giving money to select colleges specifically to develop mainframe-centric curriculum.