[Info-vax] C99 stuff (Re: The Road to V9.0)

[Bob Gezelter]

On Sunday, June 9, 2019 at 3:19:57 PM UTC-4, Dave Froble wrote:
> On 6/9/2019 2:46 PM, Stephen Hoffman wrote:
> > On 2019-06-09 18:11:45 +0000, Robert A. Brooks said:
> >
> >> As you likely know during a negotiation, you typically need to give
> >> something up to get something.  I cannot imagine VSI wanting to give
> >> up something for the right to open-source (or otherwise distribute)
> >> the HP-owned VMS source.
> >
> > As HPE exits the OpenVMS business, HPE has made what revenue that they
> > can with OpenVMS.
> > No new HPE patches starting in less than two years, and best-effort
> > support past that.
> > That's a seemingly reasonable time to transfer full source rights to the
> > exclusive licensee.
> > That'll probably be the last hunk of cash HPE can acquire from VSI, if
> > HPE are inclined to ask for it and not simply transfer it.
> > Without those rights, HPE affixes an anchor on the future options
> > available for VSI, and for any potential acquisition of VSI.
> > As for why transfer sooner, rather than later?  The more revenue VSI
> > makes, the more that HPE might be inclined to ask for.
> > All sheer speculation, of course.
> >
> >
> 
> Yeah, what he wrote ....
> 
> In general, there may be some things where the restrictions effect 
> everybody negatively.  No specifics, just a general comment.
> 
> For an example, if VSI wasn't interested, but another party was 
> interested in taking a language front end and adapting it to LLVM or 
> something else, that doesn't seem to be so easy.
> 
> I guess VSI could have some things done "under contract", which they 
> could then make available.
> 
> And, just to emphasize, anyone who has read my past posts should realize 
> I'm not one of those fans of open source.
> 
> -- 
> David Froble                       Tel: 724-529-0450
> Dave Froble Enterprises, Inc.      E-Mail: davef at tsoft-inc.com
> DFE Ultralights, Inc.
> 170 Grimplin Road
> Vanderbilt, PA  15486

David,

The most important point is that Open/closed source is a poor taxonomy.

OpenVMS is best described as "visible source". There has been a long history of fairly detailed technical information published to the general public, such as the documentation set and the IDSM. A truly closed source model would have zero internals information available, to licensees and the public alike.

There is also a wealth of information provided to all licensees as part of the system distribution which allow and enable access to privileged and internal interfaces. While this information is provided under license, it is by no means "open" source of any kind.

The better model than Open/closed, or even a linear world is some form of hyperspace matrix with a variety of axes. I did not say it was conceptually simple, but it would be accurate.

Access to the LLVM-GEM interface would be something new, but it is in more of the spirit of the existing header files which are supplied to interface to other system components. If there are issues relating to patents, they can be identified and resolved (Most of the DIGITAL patents have expired at this point, it being 20 years since the COMPAQ acquisition).

The key to enabling such access is the process of identifying the relevant issues and making the appropriate argument, based upon the license and other agreements. Then VSI can communicate with HPE and others to ensure things are on a solid footing.

It is a legal question, not a question of philosophy.

- Bob Gezelter, http://www.rlgsc.com