It will also be interesting - in the widest possible sense of the word - to see how the interactions between altnet and 2017-era Microsoft play out as compared to the 2007-era Microsoft. As the saying goes, watch this space :)
I work with Java and .NET every day and the day our customers UNIX servers will run .NET is still very far away.
Why's that true for the customers you mentioned?
Java is a complete platform, not only CLI and Web, has application servers like Websphere that offer a full server OS abstraction across the cluster and run in several flavours of enterprise grade UNIX and mainframes, not only GNU/Linux.
Also every database worth using in the enterprise space has JDBC drivers, whereas good luck getting ADO.NET drivers or mainframe connectors for .NET Core on those platforms.
And many of the .NET libraries don't run out of the box on either Mono or .NET Core, so unless they get ported, they won't be available.
Oh, and as far as I am aware you still don't get performance monitoring tooling like the JetBrain products talking to .NET Core.
Also they have legions of replaceable programmers used to Java, unless the IT department changes their infrastructure and spends money in trainings, it won't happen, even if .NET Core already had feature parity with Java.
When you get a list of solid points about why Java won't be displaced and the response is watch this space, it reminds me of the folks saying "this is the year of the Linux desktop" without pointing to actual improvements coming to life.
The "new" .Net CLR really only has a compelling story if you are already a .Net shop. Nothing in the new Core and Standard implementations gives you something new if you are outside of it.
Hopefully I don't need to point out the very well publicised and ongoing efforts to revamp the .NET CLR and peripheral items to run on non-Windows systems.
The efforts to revamp .Net CLR to run on non-Windows systems is of little value to shops that are not already running .Net CLR. At best it is removing a negative for running .Net CLR. There is no net gain for a Rails or JVM shop for example.
The gain is only for shops that already paying the Windows tax to get .Net.
As for SQL Server, it is better than last time I checked. Do you have data about production loads on GNU/Linux?
Regarding .NET, that is only for 100% Windows stacks, even if they run across VMs on Amazon or Azure instalations.
I am yet to see a single RFP asking for .NET deployments on GNU/Linux.
They don't have any reason to switch to GNU/Linux VM instances instead of Windows ones, the infrastructure costs are a tiny portion of the overall project costs.
For instance NuGet was Microsoft's project which completely ignored the fact that OpenWrap (an OSS project) had existed and was gaining momentum. But then NuGet came along and obviously the Microsoft loud speaker basically killed OpenWrap over night. That is just one example - but there are dozens of others.
ALT.NET was also about the social aspect, such as Twitter, meetups and non-Microsoft-owned .NET conferences.
I don't see the need for any kind of similar movement right now. Microsoft is firing on all cylinders and I believe it will only get stronger, wiser and bolder now.
It isn't long now before the .NET CLR will replace the JVM as the open source standard on which enterprise apps are built (and more) - watch this space!