[Info-vax] The dangers of being too emotionally involved, was: Re: How about RdB for x86 VMS?

[Stephen Hoffman]

On 2014-09-04 00:37:04 +0000, Simon Clubley said:

> On 2014-09-03, Stephen Hoffman <seaohveh at hoffmanlabs.invalid> wrote:
>> 
>> What is going on here, folks?
>> 
>> VMS is looking pretty good right now.  This for the first time in a  
>> long time, too.
> 
> What's caused you to suddenly change your tune from your previous 
> (accurate) comments about how much work is required to make VMS viable 
> to newcomers in today's world ?

You presume I've changed positions here?  You'd be wrong.  Very wrong.

If you're expecting support statements or announcements ahead of the 
presentations at the boot camp, you're going to be disappointed.

Expecting published support statements from third-party vendors at this 
early stage is somewhere between premature and overly optimistic, in 
the most charitable of phrasing.

Expecting that VSI and Oracle are not aware that customers will want 
and need Rdb?   VSI needs Rdb for their own internal operations, or 
they're looking at a database port.

But if you really do desire official statements on VMS or on 
third-party products, contact VSI, or the vendor.  Directly.

It's questionable whether most folks need these official statements 
with any urgency, too.  Beyond some nice polite messages to assuage the 
inattentive, and maybe fodder to justify a trip to the boot camp, that 
is.    At this juncture, most folks using VMS are probably best waiting 
and watching.   With the exception of some key third-party software 
developers, most folks are not going to be working on ports to x86 for 
several years, even in the best case.

VSI will establish series of development goals.  Boot camp 
presentations.   Paramount among the engineering goals: get V8.poulson 
shipping.  After V8.poulson ships, get x64 shipping.   There'll 
undoubtedly be patch roll-ups and some updates for security patches and 
some "housekeeping" work in V8.poulson and x64, and there may be some 
features targeted for specific (large) customers, but I'd be surprised 
if there were much significant new or general work in either V8.poulson 
or in the VMS x86 port.

VSI has a huge pile of work ahead of them, now and beyond V8.poulson 
and VMS x86.   Using various other platforms regularly in addition to 
VMS, it's easy to get a decent idea of how much work is in front of VSI 
here, to bring VMS to competitive, though that depends on what markets 
they're aiming at.   The Apple Xcode IDE and llvm/clang/swift is so far 
beyond LSEDIT, it's not even really comparable.  But I digress.

I'm among the folks that would like to see VSI succeed.  Whether they 
will succeed, I don't know.  But they're clearly funded, and they're 
trying.   Any operating system project has to be a very long-term 
investment, and not is without financial risk.  Whether the investment 
and the development effort will pay off for VSI, I don't know.

Are there things here that concern me about VMS and about VSI?  Sure.   
 Fscking support statements for a release that's — again — probably 
three years out and in the very best case?  Those statements are not 
among my concerns.  I'm going to allow VSI the benefit of the doubt 
here.

As for the current status of VMS looking pretty good right now, five 
weeks ago, VMS was supported by HP and bugs fixed, but not being 
enhanced, with the most recent VMS release from 2010, and with HP 
events and presentations were increasingly covering on user migrations. 
  In the past five weeks, an additional company with a new team of 
developers are starting work on VMS, and with a roadmap for new VMS 
releases and updates and a port to commodity hardware.  That's a very 
positive change.   VMS won't really be back until after V8.poulson and 
VMS x86 ship.  And work.  But things are looking better, and in just 
the past five weeks.





-- 
Pure Personal Opinion | HoffmanLabs LLC