[Info-vax] decc$stat() variants on VMS.

[Stephen Hoffman]

On 2016-06-29 23:16:33 +0000, lawrencedo99 at gmail.com said:

> On Thursday, June 30, 2016 at 7:50:26 AM UTC+12, Stephen Hoffman wrote:
>> Give folks something to move forward to, and to look forward to.   When 
>> you have to make a choice, make that trade-off facing forward.   Not 
>> backward.  Preferably — with C — toward C11 and POSIX and better 
>> compatibility.
> 
> How about building a VMS API layer on top of a Linux kernel? That would 
> giv0e you immediate access to a whole lot of newer hardware, but it can 
> also run on your existing Alpha and Itanium hardware if you want. It 
> should probably take less work than maintaining your own entire OS.
> 
> Like WINE does the same thing for bringing Windows APIs to Linux.
> 
> What would we call it?

Something similar to FreeVMS?  http://www.freevms.net

An overhaul of OpenVMS more than a little way toward a wholesale rewrite?

Something that would require careful packaging to stay compliant with 
the Linux licensing?

Might want to do some reading on a DEC advanced development project 
from ~20 years ago, too: http://labs.hoffmanlabs.com/node/361


>> (Going toward OO APIs is probably a bridge too far, though.)
> 
> That option would be available at little or no extra cost.


That "little or no extra cost" for an OO framework for just the central 
parts of OpenVMS APIs would be a design and an implementation I'd be 
interested in learning more about.  Because any approach seems to 
involve a whole lot of work rethinking and wrapping the existing APIs, 
whether for the overall design or around lower-level details such as 
memory management and garbage collection.






-- 
Pure Personal Opinion | HoffmanLabs LLC