[Info-vax] VMS port to x86

[Keith Parris]

On 5/24/2012 11:35 AM, JF Mezei wrote:
> The only aspect the community can do is convince HP to announce it will
> open source VMS on the day it announces the end of the line for iA64 VMS
> and HP-UX.

OpenMPE was unable to convince HP to open-source MPE code after its EOL 
announcement. 3rd-party support organizations were allowed licenses only 
to view the MPE code, not to modify or redistribute it. I suspect there 
would be no better outcome with OpenVMS.

And based on the feedback I've heard at user group events in the past, 
some OpenVMS users, particularly those concerned about security, don't 
want the OpenVMS code exposed to everyone, including potential hackers. 
Security by obscurity isn't the best security, but it does make things 
harder for most potential attackers to a degree that makes a difference 
in the real world.

And even if the actual VMS code were released to Open Source, what you'd 
have is a kernel written in mostly Macro, with some Bliss and C, 
utilities and RTLs in mostly Bliss, and no Bliss or Macro compiler for 
any other architecture. Perhaps the community could write Bliss and 
Macro front ends for gcc or something, but it seems to me that would be 
the only way to make the existing OpenVMS source code usable.