HP/DEC/Compaq

Axiom does not mean what you want it to mean...

I think it's a very sad announcement although to be honest I think VMS has been under the chop for so long it was inevitable. For it to last until 2020 is really quite amazing when you think about it. I say this given the lack of features that really are required to do business in the modern world. I am fully aware of the benefits of VMS (clustering, security etc.) but it really hasn't had any serious new development in at least 10 years. When Stephen Hoffman can intelligently rattle off a list of show stoppers for modern development you really should be listening. The sad fact is that without substantial investment and development effort the operating system really is stuck in the 90's.

From a hobbyist perspective I would really like to see the hobbyist program to continue. I love the environment, I celebrate the diversity, documentation and stability. It makes a great juxtaposition from virtually all 'modern' operating systems, and I get a lovely, warm, comfortable feeling whenever I'm at the dollar prompt. Unfortunately the diversity is really the Achilles heal when trying to leverage Open Source software. The fact is that the majority of new software development these days relies on a substantial amount of open source software. Windows and Unix are close enough to have the gap bridged, but OpenVMS has never gotten to the point where convergence was realistic in this perspective.

My 2c worth!

I just realized that this is another milestone - the last bit of DEC to be buried. Compaq killed off VAX, HP killed off Tru64 shortly after buying it and Alpha somewhat later (those pesky customers - they kept buying it instead of quietly going to Integrity). StorageWorks was dropped, now OVMS.

This is the bullshit that passes for informed comment these days ? Urban, people are exactly the same as they have been for 100,000 years. They need exactly the same things. The rest of this doublespeak is just a bunch of gobbledygook. "Modern software development ..." who gives a rat's ass ? You could take all software development and shove it off a bridge and we'd be way better off.

Software hasn't done shit in twenty years. It's worthless.

Good points about open source vs. proprietary. Using the IBM example of supporting their mainframe platforms, it's obvious that HP could have kept VMS as a major long-term OS choice. But IBM still does active development on z/OS, from what I understand. HP has not been investing money in keeping VMS up to date with the latest hardware. They could if they wanted to, but the longterm vision just isn't there.

But I'm pretty sure pedantic does.

HP finally put an end to VMS, in other words: VMS an end of a waste of my life. I guess I should be thankful that they didn't drag it out too much, because it was (indeed) obvious that things were going nowhere. Even Tru64 UNIX, butchered many years ago, is in several areas more up to date than VMS currently is.

I'm so extremely glad I sold nearly all of my “AXP” and “I64” systems off, except a few Multia/UDB s (but luckily there's still, say, NetBSD/alpha).

To conclude this post, one big fuck you to my favorite scumpany HP.

Oh snap!

It's just that some of us come from countries that are, or were up to some recent point, dictatorships. And we got a chance to experience, first hand, that them dictatorships are far from efficient, as far as systems of governance go of course. Thus the axiomatic business...


Anyhoo, carry on...

Sorry, but I totally disagree with this statement. I personally would see the architectures and protocols that we have to use in web development substantially different but how you can argue that software hasn't had a significant impact on the 'average Jo' is beyond me.

With regard to VMS lacking required features - there may be niches such as embedded development where it might not be so, but one feature for example, Unicode, is now glaringly lacking from VMS compared to other current operating systems. As far as I'm concerned the lack of Unicode support is a serious hindrance to developing modern software. The company I work for runs their online application in seven different languages, this would be next to impossible using VMS (in terms of interoperability) except for Java support. However, Java basically turns it's back on virtually all of the good stuff in VMS which subsequently negates its' benefits.

What annoys me personally is the announcement that VMS will not be ported to Poulson. Having been a laughing stock with regard to performance since the initial development of Itanium this generation of the processor appears to finally be delivering on some of the promises of the architecture. VMS won't see this benefit.

It's not that there haven't been noticeable impacts, it's that they haven't been ones that are really all that great. It's cool that we've got forums like this, and we can send email across the world many times faster than it takes to mail a letter, but overall were we really that much worse off in, say, the '70s or '80s for not having "modern" software? Seems to me we got along just fine back then.


That's what really burns about this - HP having been trying to bury OpenVMS since they acquired it, this is just them finally getting it positioned properly in the grave. It's all politics...there ought to be a law that a company that buys out a product line is obligated to support it indefinitely, then maybe we wouldn't get so many companies that acquire something for the purpose of squashing and/or "buying" its customer base.

Computing technologies are quantitative in nature, which is why I suspect these sort of qualitative/subjective debates usually lead to nowhere. E.g. From my perspective, there have been some fantastic advances and some great technologies have happened quite recently, and are happening as we speak.

We tend to mistake our first contact, or the mastering of a specific implementation or process, with the actual essence of the technology itself. I'll assume you came of age when home computers were starting to be adopted. I bet you that if you interact with a young kid these days, a person that has grown in a world where the internet has happened and it is taken for granted, you two will have a completely different mental concepts or images for what "computer," or what "great" in the context of measuring a technology mean to either of you.

Again, though, I'm not saying there haven't been some spiffy things done, I'm saying that the Dark Ages didn't extend through the 1990s. The fact that a kid who's grown up with PCs being commonplace fixtures of daily life might see it that way doesn't make it so.

So, what medium would we be using to have this conversation across the globe if we hadn't progress? My guess is the closest thing would have been DECnet and VMS notes. Keeping it on topic of course

Hey, those worked just fine...heck, VMS Notes is still available on a lot of systems

That's not what I meant, at all.

I think the point being made is that the quality of software hasn't matched the great increases in hardware. For example, the Web is progress in some way, but it's a pretty ugly solution, made from poorly conceived "standards" all held together with duct tape. Alan Kay also has this view:

http://www.drdobbs.com/architecture-and-design/interview-with-alan-kay/240003442


I think, though, that there has undeniably been some progress (consider the Internet and networking in general). We can also look at the state of Unix systems. Back in the 80's, scripting would have been done with Bourne shell and awk. These days we also have very popular interpreted languages like Python and Ruby. These can fill in the big gap between C and simple shell scripts, and help us avoid the ugliness of Perl.

Languages like Go also show some real progress made in concurrent programming, and the same might be said of Erlang as well. There is some substantial progress, but it happens fairly slowly, and most people are still happy to reinvent the wheel for the Nth time.

The big failing, in my opinion, has been in mainstream graphical programs like web browsers. It seems that there is still no elegant way to create a GUI application, no clear set of primitives from which they can be composed in a modular manner. GUI applications are still ugly, and little progress has been made on these.

A sad day...

Spent many an all-nighter back in 1996 optimizing a 3100/90 (or 95/98?) on behalf at the Thrifty Foods HQ in Victoria, BC... happy days.

Before that at Liverpool Uni.... a year of VMS, and even before that three years of VMS, RSX-11 and RT-11..... back in the early 90's they were ancient then

Ended my VMS experience by selling my 3100/38 for $200 back in '96 as well.. we dumped the uVAX-II (which was donated to me beforehand by the Uni... thanks Mike Senior... respect.. must have been the only student with a MicroVAX in my dorm room!)

HP might have nuked OpenVMS, but its parent, PDP-11, is still spry and powering GE nuclear power-plant robots and will do for another 37 years .

I think David Cutler isn't retired yet, and he singlehandedly designed and wrote RSX-11m according to some. I doubt they could compete with what Microsoft pay though