[Info-vax] VMS on x86-64 timescales, was: Re: x86 Update 4/22/19

[Jan-Erik Söderholm]

Den 2019-04-30 kl. 16:46, skrev Dave Froble:
> On 4/30/2019 8:22 AM, Simon Clubley wrote:
>> On 2019-04-29, Dave Froble <davef at tsoft-inc.com> wrote:
>>> On 4/29/2019 9:46 AM, Simon Clubley wrote:
>>>>
>>>> That's possible, if you mean late 2020, but I doubt many will
>>>> use it in production for at least a year after that until the
>>>> initial teething issues have been worked out.
>>>
>>> How else does one work out initial teething issues other than to use the
>>> product?  It's not something that just happens over time.
>>>
>>> My plan is to do some serious testing as soon as possible.  Yeah, fine,
>>> plan all I want.  I know that the real testing happens when you actually
>>> attempt to use the product.  At least it's always happened that way in
>>> the past.
>>>
>>
>> Planning to test something is not the same thing as being willing
>> to use it in production. :-)
>>
>> Simon.
>>
> 
> Well, if in testing serious problems are discovered, then no, one doesn't 
> use the product.  But usually, the final testing is "use in production".
> 
> Nor will testing be just an abstract project.  The goal will be to use the 
> systems in production.
> 
> So I don't really understand your comment?
> 

Maybe semantics, but one could call the activites done in a test
environment for "tests". Then many have a separate environment (still
not production but very similar) where "User Acceptance" activites
are done together, and in many cases, by the end users.

When going to the true production systems, things shoulld work but
one obviously has to be prepared if something pops up. In that case
it is usually some functionallity that was missed, not any true
technical issues, that should have been caught earlier...

And yes, sometimes "tests" are just POCs and never meant to be
part of a production environment.