[Info-vax] OS specific APIs, was: Re: September 6, 2016 - new Roadmap and State of the Port updates now on VSI website

[David Froble]

Simon Clubley wrote:
> On 2016-09-10, David Froble <davef at tsoft-inc.com> wrote:
>> Personal opinion.
>>
>> I think that when you implement an environment on top of another environment, 
>> then you're no longer really running executables under the lower (not sure what 
>> term) environment.  In this case, if special stuff is implemented in C, then by 
>> using it you're no longer running on VMS.
>>
> 
> And what if that special stuff is needed to get code written to run under
> another operating system running on VMS as well ?
> 
>> I also think that if you're going to claim that you're running executables on 
>> VMS, then it should always be VMS services used to do things, not some language 
>> specific replacement of a VMS service.
>>
> 
> So IOW, what you are saying is that any code imported from, say, Unix
> should be re-written to use VMS specific APIs instead and that people
> should go to the effort of maintaining a seperate VMS version of the
> code ?
> 
> Also how would you handle new versions of the main code base; would
> those same people have to manually import changes in the main version
> of the code into the VMS specific version ?
> 
> Good luck with that... :-)
> 
> Simon.
> 

First, let me state that I'm approaching this from a software architectural 
perspective, not so much a practical perspective.  You may have noticed that 
some have claimed the entire C mess on VMS is a horrible kludge?  The reason is 
most likely because of the lack of maintenance of VMS for the last 20 years.

As for other OSs, I don't use them for the things I use VMS for, so for me it's 
not an issue.  Now, if I needed software running on another OS, my choice would 
be to just use that other OS. Why would anyone want to go to the extra effort to 
port something that doesn't fit well on VMS?  Doesn't sound reasonable to me.

If a particular application is needed on VMS, my choice would be to implement it 
on VMS, rather than pounding a square peg through a round hole.  Case in point, 
TCP/IP.