[Info-vax] Out with Hurd, in with OpenVMS

[John Wallace]

On Aug 22, 5:47 pm, "John Reagan" <johnrrea... at earthlink.net> wrote:
> "JF Mezei" <jfmezei.spam... at vaxination.ca> wrote in message
>
> news:4c7036e8$0$10446$c3e8da3 at news.astraweb.com...
>
> > re: DEC's Compilers being more expensive because they were better.
>
> > Sorry, but there is no way that VAX-C's superiority to Microsoft's total
> > development environment warranted a 10 times cost difference ($6000
> > versus $600).
>
> I seriously doubt anybody paid list price for a compiler license.  Back in
> the 90s I saw a report showing that compiler licenses averaged an 80%
> discount over list price.  They were often just given away as "freebies" to
> close a big hardware/support deal.
>
> John

Thank you JR for pointing out how ridiculous at least one of these
claims was. I've got rather bored with pointing out real facts rather
than distant and selective memories. Yes much of the time the pricing
was stupid and yes it was probably too stupid for too long, otherwise
we wouldn't be having this discussion. But moves in the right
direction were happening. Just not soon enough.

But here are a few more verifiable (and quite possibly selective :))
facts, from my DECdirect UK Software catalogue, February 1994 (because
it's the one at hand).

DEC C for VMS system-wide list prices start at £2000 for a small
system and go up according to "points" (who remembers them?). "System
wide" licencing was not the only option (and even if they were, they
were priced according to "points" on a system or cluster), many
products including compilers were offered on a "named user" or
"concurrent use" licence basis.

A single named user for DEC C for OpenVMS was £800 list, and the
debugger is included with the OS at no extra cost.

Obviously if you have lots of developers there's a break-even point
where system wide licensing is more cost effective. Wrt development
environment: source code manager (CMS) and a build manager (MMS) were
extra, just as they still are with Studio. And (as JR already said),
you had to be quite unlucky to pay list. By the time Alpha arrived,
low end list prices existed largely to allow resellers to make a
margin on passing an order through their books even though they hadn't
actually touched the order in any meaningful way. Cutting the
resellers out of the loop for commodity orders (the DECdirect
approach_ would have allowed even more sensible pricing, and indeed
worked well when DEC UK tried it. But it (obviously) upset the
resellers. The Wintel world has this right; buy Wintel and everybody
benefits (apart from the folks paying the bill).

Anyway, the real target for price comparisons back then surely
shouldn't be Wintel PCs, surely it should be Apple or Sun/IBM/etc? OK
Sun aren't doing so well any more (no surprises really), but the
believers in the Church of Jobs still seem quite willing to pay over
the odds for what is now basically industry standard x86 hardware with
a custom logo. This despite Apple kit having an OS that doesn't
natively run Windows apps, but it does do everything the faithful
need, without being a prime virus-transmission vector. How did that
work for Apple and not for DEC?

In a less fashion-led more engineering-led sector of industry, you
could say that National Instruments are to PC-based C+I (control and
instrumentation) what DEC used to be to minicomputer-based C+I. NI
prices are **outrageous** in comparison with their more focused
competitors. But NI do OK and are the default choice for that market
sector, same as VMS once used to be the default choice for its market
sector. How did that work for NI and not for DEC?