[Info-vax] Happy new Year !

[JF Mezei]

Richard B. Gilbert wrote:

> "A stern chase is a long chase!"  Especially so when you are 999 million 
> systems behind!

VMS used to be ahead in terms of office automation desktops. And used to
be way ahead in number of networked computers/users.

Remember that if VMS were able to capture 1% of the market today, it
would represent incredible growth for VMS. And such growth would give
VMS some much needed energy to continue to grow.


> I'm not going to hold my breath!

At the time Compaq bought Digital, VMS still had a chance to be
resurrected. In fact, the short lived "renaissance" where VMS was
allowed a tiny bit of marketing did show that it was possible to grow
VMS significantly.

One of the arguments used as an excuse to murder Alpha is that by moving
VMS to an "industry standard commodity CPU", it would enable a wider
breath of applications from desktops to data centres. (which had been
the original goal of IA64 back in the 1990s).

Initially, we were told to wait because once VMS was commercially
available on that IA64 thing, HP would start to market it as one of its
own products. It didn't. And that was the last reasonable window of
opportunity for a true VMS renaissance.

Remember that after the port to IA64 was done, HP began to downsize VMS
engineering.  Had it maintained the higher staffing levels, they would
have been able to push ahead with plans to make VMS more compatible with
Unix, and had they been able to complete this quickly, then all the
desktop Linux applications could have easily been made available on VMS.
And if done right, the #ifdefs for VMS could have been incorporated into
the main open source code, making every new version of the product
quickly available on VMS.

They did manage to port Mozilla to VMS (and all the middleware needed
for it). The problem is that they didn't have the resources to "finish
the job" by having all of the customizations sent back to the open
source   code so that it would then be easy to build the apps on VMS
whenever new versions come along. But the port of Mozilla was a proof
that it was possible to build modern applications on VMS.

Yes, it was possible to give VMS a new life, but that would have
required a nudge in both marketing and engineering resources. Neither
happened. Plans to make VMS "unix compatible" remain, but they are now
just pipe dreams still on the roadmap but without the quick delivery
needed to allow easy porting of Linux apps to VMS.

IA64 on the desktop was officially abandonned in early 2004 at roughly
the same time that Intel capitulated to market pressures and agreed to
produce what the market wanted: a 64 bit 8086.


In my case, I held my breath until 2004. HP had a chance between 2001
and 2004 to show that it would give VMS a chance. It didn't.

Between 2004 and 2009, it was a case of "not holding my breath, doesn't
look good, but still give HP a chance". As I recall, there was a speech
by Livermore in fall of 2008 which sealed the deal when she would not
include VMS in the "actively develop" and specified that they would
"support the VMS installed base".

And of course first week of january 2009, I learned that VMS engineering
was being disbanded.


Having said this, despite all the criticisms of the TCPIP Services
product, VMS had made great leaps forwards in gaining the functionality
of Unix. And now that I am learning Unix, I am also realising that the
TCPIP Services did have a few advantages and design improvements over
standard Unix stuff. Heck, the TCPIP> SHOW SERVICES display is
certaintly much better status of your system than LAUNCHDCTL LIST