Hewlett-Packard has changed its direction on OpenVMS. Instead of pushing its users off the system, it has licensed OpenVMS to a new firm that plans to develop ports to the latest Itanium chips and is promising eventual support for x86 processors.
Great news for OpenVMS, and a great move by HP.
I’m not just saying this because I prefer Free Software, I’m saying this because having source code, a community, and a culture is what kept the unix family alive and led to its modern incarnations like Linux and the BSDs. VMS had none of that, which is a large part of why it’s basically a dead OS compared to *nix.
I have to disagree with you there.
Back in the days of DEC we had regular DECUS symposiums.
There were lots of free Software available for it.
VMS wasn’t as closed as you might think. The source was available on Microfiche and even on a 2400ft reel of Magtape. That all died a death when Compaq took us over.
IMHO most Unix systems are the walking dead. Linux has taken over in many cases.
{Disclaimer: I worked for DEC in Reading from 1978 to 1999}
Edited 2014-08-01 11:35 UTC
OS X i unix certified though. On the desktop i believe OSX have a larger market share than linux.
Disclaimer, i think personally OS X is vile to work with as a seeing impaired person, especially now that everything is going flat over there as well.
Edited 2014-08-01 12:31 UTC
What: “larger market share than linux”.
Something that can be copied freely does not have a “marketshare”.
It is impossible to know how many people use linux.
What a ridiculous thing to say. Of course it can.
Market share? Linux doesn’t participate in a market.
Of course it does. It just happens that its price point is well below that of the competition. Free market economy at its finest.
One wonders from where / from what train of thought the two idiocies above came from…
???
VMS is a server OS, so this is basically out of topic.
Could you freely modify and distribute the source? Having the source available under a restrictive license is not enough
This article is not about open sourcing! It is about licensing an extremely high-end OS to make it work on extremely high-end hardware so extremely demanding customers can buy extremely expensive hardware and software that works extremely well.
Expecting this to become Open Source and free-for-all is extremely naive.
“Having the source available under a restrictive license is not enough”…well, it is actually EXACTLY what that partner wants because now they can earn back the money that they paid for the licensing
Did I make that extremely clear?
Edited 2014-08-01 14:57 UTC
It is not always necessary to open source, in this case certain patents make that impossible.
Considering the stability and security of OpenVMS one would recommend using it for it is far better then any Windows or Linux (I use both).
Although hardly known, Samba has some roots in VMS.
And as far as I remember there are more very basic technologies that find their roots in VMS. Not to forget the windows-filesystem NTFS.
I hope it will revive then I have never used a better OS.
Are there likely to be any patents in OpenVMS that are still extant, or at least not already very close to expiring?
I think it is well known that many technologies in Windows NT have their roots in VMS (so SAMBA and NTFS should come at no surprise).
There is also the VMS -> WNT increment letter joke (just as IBM->HAL).
I think It is common knowledge that Dave Cutler led Windows NT development, and his background is as VMS as it can be.
NTFS has nothing to do with OpenVMS.
http://windowsitpro.com/windows-client/windows-nt-and-vms-rest-stor…
NTFS is an evolution of OS/2’s HPFS. It has nothing to do with OpenVMS.
As you can read in the mentioned article. It primarily emphasizes the “NT” kernel sharing many similarities with VMS. On the other hand, it does not mention NTFS in this context. I should have written this in my previous post in order to help assigning the article’s content properly. Maybe it’s not known well enough where “NT” and “NTFS” have their roots, so a little confusion did arise… 🙂
Read the article:
“Most of NT’s lead developers, including VMS’s chief architect, came from Digital, and their background heavily influenced NT’s development.”
I happen to have had a conversation with the designer of NTFS Dave Cutler in Holland during a DECUS meeting.
If you took the trouble to search in google you could have found the following:
“They (Files-11 and VMS) are “similar” because they had the same designer Dave Cutler, originally from Digital Equipment Corporation. Dave later went to Microsoft.”
and
“NTFS owes some of its architectural design to Files-11 used by VMS”
If NTFS shares some features with Files-11 it’s because the designers of HPFS chose to copy them, not because Dave Cutler designed NTFS.
In other words, any similarity with VMS is indirect, not direct.
Huh? That article does not seem to mention anything on NTFS.
BTW, for those interested in knowing more about NT’s lineage I recommend G. Zachary Pascal’s book “Showstopper.” It’s basically a novelized collection of 1st hand accounts of the team that developed NT. Most shared characteristics between NT and VMS are, at best, coincidental.
http://books.google.com/books?id=o2IkAwAAQBAJ&lpg=PP1&pg=PP1#v=onep…
Exactly what I was saying. 🙂
I wanted to emphasize that many similarities have a reason: involved persons and adopted design concepts. I think the article I mentioned illustrates this quite well. This has also been expressed in other comments here.
Edited 2014-08-04 21:22 UTC
Software patents and people who enforce them can take a flying fuck at a rolling donut. The whole concept is abhorrent.
Why it’s not enough?
You can modify the source to make the software fit your needs, and you can distribute patches to help thy neighbour ™ acquire your modifications (by applying them on the original source he also got from DEC).
How is having redistribution rights to the entire source, aka DEC not having exclusive redistribution rights a precondition for freedom or whatever?
Edited 2014-08-01 17:50 UTC
I am agreeing with grandparent. If I make a modification to the Linux/FreeBSD/etc kernel I can freely distribute it. That´s the reason why they are alive. I am pretty sure you could not legally distribute your patches to you neighbour without paying a fee because it would be a derivative work covered by the license.
Since no one answered your question… No, you could not freely modify and distribute the source code. It was provided primarily for reference by those developing privileged code that had to interact with the kernel.
Also, not all the VMS source code was available for review by customers.
We used to have that stack of fiche. It was, oh, about 4-5″ thick. Kinda heavy too.
Basically, they microfiched program listings (that is the listings the compilers/assemblers created).
I did not look at them much, we didn’t need to. What little I saw was all Macro code. But I think VMS used a lot of BLISS too.
Even source code you can’t change is useful. I often refer to source of open source projects, yet have no need to modify them. I use them as augmented documentation. It’s great for hunting down obscure error messages.
Mind, having it on micro fiche — not really useful. If it were on tape, and thus on our system, i’m sure I would have spelunked a lot more.
We did spend a lot of time in the Wall Of Orange, trawling the manuals for tidbits.
This. This right here.
Back in the late ’80s early ’90s, I was writing device drivers for both VMS and Ultrix. When I had a question about how VMS worked, I just looked it up in the fiche. When I had a question about how Ultrix worked, I had to go by rumor and innuendo, because I was not attached to one of those few institutions that had a Unix source license.
Things got a bit better when NetBSD happened, but Unix documentation has never been up to VMS snuff.
I have an interesting thought: What about operating systems derived or inspired by the original VMS that have been “created” (more or less, probably less) outside of DEC? I’m talking, for example, about SVP (MOSVP) for the robotron RVS K1840 supermini build in the German Democratic Republic during the late 1980. The hardware itself was “derived” from a DEC VAX 11/780 (which “KoKo” got two machines through “secret trading channels”) and “re-implemented” entirely with home-grown hardware components. The software looked and behaved like VMS, but it was “per definition” something else. Still it provided 100 % compatibility which was mandatory in order to run commercial software designed for the VAX. There’s hardly any knowledge about where the software “templates” came from, and other eastern-block countries also had their VAX clones, running original VMS or “native” solutions. It’s quite possible that program code, printed, on mangnetic tape, on film, or disk packs, somehow came across the “iron curtain” to furthermore “inspire” the “independent” development.
In the final product, system messages and copyright headers were probably “adapted” to reflect the work put into the whole project, so I assume that there had to be some reels with source code involved. Now let’s just assume someone gets the source code of SVP, “(C) VEB robotron Dresden 1987 so and so” and puts it into the public domain, Would that be a usable basis for “modernizing” the code legally? (I assume it’s not, but maybe it could influence independent development…)
Providing the source with the machine on delivery was common because of the lack of commercially available software. The users could implement it according to their own requirements, which was quite common and spawned many “individualized solutions”.
Historical sidenote: The K1840 (like the original VAX) was able of running SVP and a UNIX derivate, MUTOS.
http://en.wikipedia.org/wiki/Robotron_K_1840
A MicroVAX II copy also was created:
http://en.wikipedia.org/wiki/Robotron_K_1820
Oh, and also note there’s FreeVMS:
http://www.freevms.net/
Just wondering, if it’s ever is ported to x86 will it be of any use now? Just asking, know next to nothing about the OS.
Nope, OpenVMS is as good as dead. As far as HP was concerned it had been on “maintenance” mode for a while, I think most of the development was transferred to a small team in India long ago.
This is nothing new, it’s very common for orphaned systems to be purchased by smaller “vulture” companies who try to extract whatever it is left of the carcass. In this case, most of the attraction of VMS comes from it’s relatively large base of institutional customers, which are willing to pay for not very competitive support contracts for the short/mid term, because the cost of the porting is more expensive than maintenance. For HP the revenue stream from VMS is not large enough to be worth the while, but for a smaller company it is still an attractive proposition.
Yeah, there are always talks about porting recently orphaned product X to x86 (or RISC in the old days), or whatever is the buzz platform of the day. But that is just pie in the sky marketing nonsense to get some free publicity. It makes no sense to invest on the porting effort required, when the platform has negative growth rates, a very old software base tied to other architectures, plus any porting effort to x86 that would be “attractive” to possible paying OpenVMS customers would require a very expensive/extensive certification and validation process to boot.
Of course, I could be wrong and a x86 port of VMS could surface. But IMO that’s highly highly unlikely.
Edited 2014-08-01 23:18 UTC
I think it will be much easier to those few who depend on legacy VMS environments to run legacy VMS software to simply use means of emulation (for example SimH) – so they avoid any rewrites of OS and application software and can use their required parts “as is”, because they probably wouldn’t want to change anything, as it’s not about “development and evolution”, but instead about “run it the same way as it was running for decades”. Because, when you’re going to do a rewrite, you typically target the platforms available today (commonly x86) instead of waiting for a “somehow legacy” platform (VMS on x86 native) to reach first release state in a few years, so the rewrite of applications would obsolete the need for VMS.
So it’s either “use SimH” (free) or “write new stuff and wait” (expensive), and many businesses seem to think in quarterly dimensions rather than next 5 years anyway… 🙂
MAN Diesel, VMS and SimH is an interesting example from 6 years ago:
http://www.migrationspecialties.com/pdf/SimH_MAN.pdf
To be fair, it worked fairly well for OS/2: eComstation is still going and has actual customers and users. So perhaps this will be good for OpenVMS and we’ll still be seeing articles on OSNews about it in 10 years time?
OpenVMS was a great system and while it was the grandfather of Windows NT, it also served as inspiration for user friendliness and rock solid dependability. OpenVMS was one of my personal influences for Black Lab Linux and I hope they do release it for x86 processors
Roberto J. Dohnert
Black Lab Software
http://www.blacklablinux.org
HP sees no future for this operating system. They’re not going to pitch it to the large enterprise customers. They’re going to pitch Linux on VMware running on HP hardware and associated software and services. They might even still give a nod to HPUX, but really the focus is on virtualized X86.
Companies sometimes do this, selling off product lines they want to kill to other companies. No one sells a product line in which they see a profitable future.
Thats not the case at all. Many companies do this because they still have a decent amount of subscribers but this happens when cost of production outweighs the cost of what its bringing in. Like OS/2, OpenVMS has a lot of dedicated users and enhancing and updating the stack is not a kiss of death. A kiss of death would be just ignoring your userbase.
Roberto J. Dohnert
Black Lab Software
http://www.blacklablinux.org
Frankly, as a long-time VMS user on VAX and Alpha both hobbyist and professionally at times, I don’t care about it not being open sourced. People talk about culture and community and how it detracts from it not being opensourced, frankly I find this as rubbish given the number of people I know that run it as a hobby and professionally.
I can’t wait to either apply for or purchase a hobbyist license for x86 VMS. The IA64 port proved that the built-in hardware support from VAX & Alpha weren’t necessary.
Seriously brilliant move to keep it alive given HP’s lack of interest the last few years. I hope and pray that the new owners encourage the hobbyist community as DEC/Compaq/HP have. It’s outstanding.
A question .. ever seen MicroSoft give away free license keys to Windows, Compilers, Applications etc? No.. never. OpenVMS Hobbyist does. Has done. Always will. So, stop your whinging, grab some old hardware off fleabay for cheap, or run a free emulator and go apply for your FREE!!! (as in beers) license to run this outstanding operating system.
If you enjoy Open Source, go hack around on a BSD or at worst Linux. If you want to experience a well designed (and well thought out!) system that is also security conscious system that excels at transaction processing, then step up.
Visit http://www.openvms.org for various clues and nice DCL routines.
– Al
Never say never. Microsoft has been giving free licenses for most of their products for academic (and other non commercial) users for ages.
True. But OpenVMS hobbyist gives licenses to all OS and compilers to everyone that applies as a hobbyist. Not just impoverished academics and students.
I wouldn’t say “just” given how those supposedly “impoverished” academics and students are orders of magnitude more than the pool of OpenVMS hobbyists.
Also Microsoft’s academic licenses don’t expire after just one year, which I think that was the case with the OpenVMS hobbyist licenses. Also they offer free access to patches and updates, which HP does not do for OpenVMS.
Don’t get me wrong, I think it’s great there is a possibility for people to access OpenVMS legally as a hobby (I used it when I had access to Alpha HW back in the day). But let’s not be ridiculous by over hyping something that in the big scheme of things is rather minimal. Specially considering the competing alternatives…