What I love most about the lawn mower analogy (which I apply to any company) is that sometimes it can also be useful.

I think like, a big part of this was Solaris has near zero chance of gaining headshare among people who need Unix. Solaris has value for customers with vendor lock in, and its a declining revenue source which can be ridden out for a decade or two, and then discontinued.

I don't think that awaits Solaris.

The Rdb database for OpenVMS is still getting updates, and a new port to the VMS-x86 release that has recently emerged.

https://www.oracle.com/database/technologies/related/rdb.htm...

VMS has a might higher degree of lock-in than Solaris does.

Yeah, probably the consistent way to explain it is that Oracle thought they could get away with it for Solaris but the market of Linux distros is quite healthy and they don't have the power to pull the same thing so they push the other angle instead.

As the poster I replied to pointed out, the lawnmower only cares about making money.

TBF Linux put OpenSolaris on life support, with RedHat's help.

Oracle just pulled the plug.

That's exactly right; I was surprised way back when that Linux wasn't cutting into the Windows marketshare very much at all, despite all the noise and fury coming from Redmond, but it was absolutely decimating big iron UNIX.

Windows was already on low-margin consumer x86 boxes so there wasn't the same opportunity.

Linux enabled commodity hardware to eat Big Iron's lunch using that same x86. Dell and IBM certainly made a good business filling datacenter racks with their x86 offerings for quite some time.

Well world is about same as before. But it does remain tricky for people who think there is only one possible behavior for an entity all the time.