[Info-vax] https://groups.google.com/forum/#!topic/comp.os.vms/v07C_K7KzCg%5B1-25%5D

[Stephen Hoffman]

On 2020-08-17 14:39:09 +0000, gérard Calliet said:

> Bigot-ism is part of the problem. We are the users, You are the 
> Always-in-truth Company. Which is the corollary of "everything is 
> aknowledged by the majority business is The Real".
> The counter-theorems being: "the majority business trend killed VMS", 
> "the successfull ecosystems are collaborative".
> 
> But yes I do think nethertheless that the VMS ecosystem exists. And the 
> Community License is the beginning of relaunch of a collaborative 
> community (you don't have to pay for it).
> 
> The core values of VMS are:
> - LOCALITY
> - CONTINUITY
> - MAINTENABILITY
> 
> The 2 first can be summarized by the genial idea of the founders: 
> mini-computing. (departemental computing, the computer is near its use, 
> and cannot stop, clustering is to scale and garanty continuity).
> The 3rd is a consequence of the 2 first.
> 
> These values are quite the opposite of the main stream of today:
> - decoralate everything from everything by all ways of abstraction, 
> delocalize (cloud),
> - reinvent the wheel as many times as it is for good profit,
> - distant and global managering.
> 
> Not to say VMS is good, new world is bad. Just to say they are quite 
> different. And have to be supported with marketing, training, 
> development, community management different.
> 
> The way VMS is in technical revival is awesome. The way it is 
> distributed and announced, the way VSI and the community does'nt 
> constitute as a collaborating ecosystem are making the boat in danger.
> 
> The astounding situation here is that the whole world is now thinking 
> about return to more locality, to think about sustanibility, to critic 
> a modernity which forgets its roots, VMS is just totally synchronous 
> with that... and we sell it as the new gadget of Mr Steve Jobs.


I'm murky on what your statement and your suggestions might be.

In no particular order...

The trade-offs around hardware, and networking, and hardware and 
software costs, and expectations, and hardware and software 
capabilities, have all shifted. Substantially. New installations and 
new apps and app overhauls all have to reflect these shifts. We're no 
longer suffering VAX constraints on memory costs and storage costs, for 
instance. Though existing app designs often are. These shifts can and 
often do then alter the decisions around the chosen locality of 
computing, of service continuity, and of maintainability.

For a few examples of these shifting trade-offs, hardware reliability 
has increased substantially with failures less often and reboots far 
faster, and spare servers are now far more affordable and even 
commonplace. And clustering for app scale is less interesting as 
hardware servers become faster and more capacious.

Micro-mini-supermini-mainframe-super is a characterization of computing 
I've not often encountered in then recent decade. Not outside of 
comp.os.vms. Not when x86-64, and Arm AArch32 and AArch64, have become 
ubiquitous throughout most of computing, outside of IBM mainframes.

"Edge" is probably closest to departmental computing, in the 
contemporary distributed design (or design marketing) usage.

Differentiating a local or private cloud and a hosted cloud and a data 
center? I'll leave that debate to others.

As presently implemented and presently used, OpenVMS itself is closer 
to an embedded operating system; an environment where the app developer 
also manages the operating system.

"Gadget" sells short the efforts involved in design, tuning, 
communications, UI, and production.  App code alone is a small part of 
a product. Productization, production, distribution, design, services 
and support, documentation or the ability to render documentation 
unnecessary, and the rest of the efforts involved are far less often 
recognized, but necessary. VSI owns all of this, as does any other 
commercial vendor.

We can and will see echos of computing's past, but we're never going 
back. And as I've grumbled, reacting to customers' requests is useful 
and necessary incremental work, but that isn't going to get a vendor 
into a new market.

Where OpenVMS fits in 2030—and how to get there—is a job for the folks 
at VSI. This is what VSI should be and is focused on. Customers are 
certainly a part of these calculations, but VSI needs to be looking 
five and ten years out; to be positioned. And how to have products that 
will be interesting to customers in three or five years.

As for this collaborative ecosystem", I'm a little murky there.  VSI 
marketing and communications do have some opportunities for 
improvements, yes. And tiny company, huge product, pandemic, way too 
much work, too little time, etc., are all factors, too.



-- 
Pure Personal Opinion | HoffmanLabs LLC