[Info-vax] Where to locate software

[johnwallace4 at yahoo.co.uk]

On Thursday, 9 June 2016 21:00:36 UTC+1, Stephen Hoffman  wrote:
> On 2016-06-09 18:55:33 +0000, Simon Clubley said:
> 
> > Sorry Stephen, but I think you have gone a bit too far there. VMS _is_ 
> > a competently and well designed system and is damned good with it's 
> > traditional workloads.
> 
> OpenVMS was very competently designed, and was and is very well suited 
> to the era that it originated in, and to the associated applications.   
> Doc was great, great networking, very consistent APIs, etc.
> 
> In the ensuing years, OpenVMS has lost more than a little of that 
> distinction, and for various reasons.   Compatibility was certainly a 
> major reason for the complexity, and there are various other factors.   
>  Sure, there's still very nice stuff in OpenVMS underneath it all.   
> But that isn't what the apps I'm working with are really using, either. 
>  Some are — the original code and the old apps — but the new apps and 
> the app updates are increasingly using things that are just a hassle to 
> work with.  Some — like 64-bit addressing and network security — is 
> most kindly referred to as being problematic.  (64-bit was an amazing 
> design, alas with the benefits of that work skewed far more toward 
> existing applications and not the future and new applications.  But I 
> digress.)
> 
> OpenVMS app packaging and app installation and app configuration and 
> the rest are not well documented — if it's even really documented — and 
> the current tools are inadequate, at best.   That's before discussing 
> mass deployments and configuration automation and replication, too.
> 
> VSI has massive piles of work ahead of them to address and to improve 
> the competitiveness of OpenVMS.   If their revenues are established, 
> maintained and can trend up, they'll hopefully continue that work.   
> App packaging (and related doc) is one of the (many) areas that needs 
> work.
> 
> ps:   Windows has rolling upgrades now?  Interesting.  That knocks 
> another of the distinguishing features off the list.  OpenVMS apps 
> could use some help here, though that's certainly part of what's should 
> be addressed here — around product installations and upgrades and such. 
>  Migrating formats in RMS files can be... entertaining, and that's part 
> of why I've been pointing to integrated databases.   This difficulty is 
> how OpenVMS clustering got to be as complex as it is — with all those 
> shared RMS files, manually referenced via logical names — after all.
> 
> 
> -- 
> Pure Personal Opinion | HoffmanLabs LLC

Prompted by rolling upgrade subtopic, but more widely applicable:
Microsoft may claim that Windows has various different innovative
features. They are after all well known for bringing 20th century
computing technology to the 21st century world.

Some people believe MS's claims. Others have tested MS's claims
and found them wanting. In recent years, many saw little option
besides migrating to MS anyway.

Nowadays, the MS ecosystem is not the industry force it was maybe
a decade or so ago. But as already noted, VSIVMS isn't going to
be credible as a routine replacement for the Windows Server world
in general,  in the same way as Windows was not actually a
universal replacement for the VMS market in general, whatever the
vendors may have said at the time. SAP on Windows NT on Alpha, for
goodness sake?

Inside today's Windows world, however, there might just be people
that have felt the pain of forcefitting Windows into the wrong 
shape hole. These people might probably look next at a Linux.

Some select few of the ex-Windows customers, e.g. those who
aren't horribly dependent on platform-specific stuff which
VSIVMS doesn't yet have, and/or those who remember when
computer systems used to Just Work(tm), *might* be persuaded
to look somewhere other than Windows and Linux, for the
specific requirements of their server class systems. Perhaps.

VSI can hopefully find enough of these folks, in addition to
the installed base and friends, to keep a small company like
VSI going. 

What kind of things need to be done to make that easier?
What kind of things tend to make it harder to address those
prospects?
And also: what differences are there between selling to
random end user organisations, and selling to outfits who
are addressing specific markets where Windows and Linux
had to be force fitted when the death of VMS was pronounced?

As regards people migrating off VMS: surely where that was
a relatively easy option they did it long ago? Soon those
people may have an opportunity to evaluate whether they
made the right decision.

Have a lot of fun.