I thought VMS development was moved to Redmond back in the late '80s ;-)
The companies that create and modernize operating systems are under the same economic pressures as the IT departments of corporations - large and small - that create and maintain their own applications atop those operating systems. And it comes as no surprise that the development of the venerable OpenVMS proprietary operating …
I thought VMS development was moved to Redmond back in the late '80s ;-)
Please don't annoy the OpenVMS trolls, otherwise the old codgers will be shaking their Zimmerframes at you! ;)
On a more serious note, the good news is at least hp are still developing the stuff for those users that want it. Personally, I'd kill it with fire, but we still do have some legacy apps running on it that just won't die, and I'm constantly told by the OpenlyVenoMouS trolls that there's a lot of the stuff humming away on spanking new Itanium kit in the financial sector. To each their own!
A rather less amusing subtext is how the far too many in the West view the idea of development being shifted Eastward. There seems to be this autonomic belief that shifting development to India somehow implies a drop in quality and the end to development due to some unspoken belief that Indians can't develop code, just imitate and support Western coders. This is verging on racism. I know plenty of Indian engineers far clever than myself and write far better code. Then again, that's probably not the best example seeing as I haven't written code for years, but you get the idea.
I'd compare it to the car industry - when Japanese companies originally made cars they were copies of Western designs. When the Japanese started importing their own designs to the West they were derided as cheap, clunky and substandard, and many "experts" claimed the Japanese could only copy Western innovations. Fast forward forty years and companies like Honda and Toyota have the car market sown-up Worldwide and are leading innovators in new automotive car tech such as hybrid vehicles. It seems to make sense for Western companies to use those Eastern capabilities.
Just when Oracle is looking for engineers to splice VMS features into Solaris, HP closes their VMS lab.
OpenVMS works to well to be considered for use. Oh, I see. Move it off shore so it can acquire all kinds of new bugs and become a more competitive product.
Phased migration time for all those vms servers before the codebase falls apart. But to what? solaris just got taken, sco chapter 11'd at last, maybe it is time to look at linux or *bsd boxen clusters.
"There seems to be this autonomic belief that shifting development to India somehow implies a drop in quality and the end to development due to some unspoken belief that Indians can't develop code, just imitate and support Western coders. This is verging on racism."
Yes, but it's also easy to understand. Many (most?) Western coders have only had exposure to the very cheapest of Indian outsourcing shops, chosen by management on the basis of their extremely low price. In general, these shops have "engineers" who are poor at communicating with their Western customers, produce shoddy work, and all too often can't even pretend to follow a requirements document. In the end, they get paid peanuts, and the customer actually spends more money figuring out and fixing the mess afterwards than they would have if they had never gone down that rabbit hole in the first place.
This does not mean all Indian engineers are of this nature, of course. I too had the pleasure of working with a handful of extremely skilled Indian engineers in my career. However, it was always in a long term arrangement - either direct hire or long term contract - in situations where cost may have been one of many, but was never the one and only, consideration.
At last. the IT media notice that VMS is still alive. Where have you been the last ten years or more?
Goodbye Sue (and the rest of the VMS community), and thanks for all the fiche.
"some unspoken belief that Indians can't develop code, just imitate and support Western coders"
Not really, it's the openly spoken belief that codebases get shifted offshore because senior management no longer cares about them, but still has contractual obligations. Like a TV exec moving a show to a dead slot so later he can justify cancelling it at the end of the series 'cos he doesn't like the producer for some reason.
If it's going to the site in Bangalore which I've visited then it is the one built by DEC. This is just a furtherance of Compaq's take over of DEC & HP.
As to the comments about Indian developers. The quality problems have nothing to do with where projects move to, it is down to moving projects. FULL STOP. Many of the people working on developing the code in Marlborough will have been working on it for 30 years. If the work is moved anywhere, all of that experience will be lost and can not be regained. Mistakes which were made in the past and then found and fixed 20 years ago will be made again, as no one working on the code will remember why it was done this way.
... it will continue to die.
Every HP customer needs to asking HP what is HP's exit strategy from Itanium?
For HP-UX on Integrity, it appears to be Red Hat on x86 c-Class blades.
For Nonstop? It is already moving to Itanium on c-Class blades. Will a port to x86 follow soon?
For OVMS? Who knows?
IBM abandoned Itanium years ago. Microsoft decommitted from Itanium with Windows Server 2007. SGI abandoned future Itanium systems before the Rackable acquisition. Fujitsu seems to be moving more towards x86, which means less SPARC and Itanium big iron. But with Oracle investing in the SPARC/Solaris side, how much longer will Fujitsu's Itanics sail? NEC still has Itanium, but how much longer?
Remember when CSI (now QuickPath), was supposed to come first on Itanium, then a year later on Xeon? Remember when Itanium was Intel's answer to single threaded performance? Remember when Itanium was Intel's answer to SMP scalability?
This is the 2nd time HP off-shored OpenVMS support. The first time they got rid of support engineers with 10 to 20 years of deep OS experience for a bunch of newbies that were working of scripts. The customer base (many of them big money institutions) went wild with the drop in support so they moved support back in to the US (but couldn't get back all of the skilled people they'd stupidly dumped.)
Now they are doing it again.
Annnnnd.... cue Sunshiner! Still crying over the death of SPARC? Upset that Fujitsu actually has a longer roadmap for their itanium servers than even their own SPARC64 boxen? Don't let the door hit you as you and your SPARC kit are removed from datacenters Worldwide!
"....Every HP customer needs to asking HP what is HP's exit strategy from Itanium?...." Why, hp-ux and Integrity are both making profits, unlike Slowaris and SPARC? You do remember what a profit is, don't you, only I know it's so long since Sunshienrs saw a profit you might think losses are the norm. And hp-ux has the most a longer public roadmap than any competing OS, which implies hp don't see the market shunning it soon. The evidence in marketshare is hp-ux on Integrity is still taking accounts away from Sun, especially in the lucrative enterprise high-end. Seems I have to remind you Sunshiners of that fact daily, can't you guys just write it on the back of your hands as your memory is obviously as shot as Sun's transactional memory attempts.
"....IBM abandoned Itanium years ago...." IBM made a big noise about dropping Itanium in an attempt to boost Power sales - it didn't work. Before they dropped Itanium, they were the second most successful Itanium vendor and sold about 10,000 Itanium servers. Think about that - that's 10,000 instances where IBM could only meet the customer's requirement with an Itanium solution, running Linux or Windows, and couldn't get the customer to take a Power or xSeries solution which they pushed much harder. Gee, I wonder about the poltiics in that decision!
"....Microsoft decommitted from Itanium with Windows Server 2007...." And now a fictional Windows release! You Sunshiners really have no clue at all about competitor kit at all. Did you mean Windows Server 2008, which has a fully-supported Itanium release? Try a bit of research, your FUD is even worse than the usual Sun male bovine manure.
"....SGI abandoned future Itanium systems before the Rackable acquisition...." <Yawn> They are still selling Itanium kit. The new SGI has yet to commit or deny support for Tukzilla, so all you're doing is mistaking your own whimsy for facts. As the recent Sunset should have shown you, relying on fantasy is not a good idea.
"....Fujitsu seems to be moving more towards x86, which means less SPARC and Itanium big iron...." <Snore!> Ditto to the SGI bit - Fujitsu still have not commited to Tukzilla or denied plans. Of course, there is a much longer roadmap availabel for Itanium than SPARC64, and all Sun SPARC is on deathrow, so it's not inconceivable that Fujitsu will save themselves some design costs, drop SPARC64 and just go with Itanium. Then again, they may stick with at least one more generation of SPARC64 as they try and mop up some of those Slowaris shops that Sun have left dangling.
"....NEC still has Itanium, but how much longer?..." Dunno, but then I don't buy their kit and so don't have a relationship with them. I'm guessing you don't either as your fantasising doesn't extend to any facts such as roadmaps or the like, which kinda brings us back to that warning about whimsy and reality from above.
"....Remember when CSI (now QuickPath), was supposed to come first on Itanium, then a year later on Xeon?...." So what? Inel saw that it worked and what an advantage it would give Xeon. Unlike Sun, Intel actually gets working kit out the door, and could bring forward the benefits of CSI to Nehalem. Meanwhile, Sun have just about given up on transactional memory and scout threads, and there is little chance of even getting them out with the uberlate Rock, let alone carrying the technology over into other SPARC designs even if Larry wants to let them spend his money.
"....Remember when Itanium was Intel's answer to single threaded performance? Remember when Itanium was Intel's answer to SMP scalability?" Itanium still is, so what is the point of that bit of blathering? It also seems to be the choice of the market, especially for shops that used to buy SPARC. Try a new record, Sunshiner, you're just sounding depressed and petulant. Sun is dead - move on! Instead of whining, why don't you get a little hp-ux or OpenVMS training and enjoy the benefits of still being hireable in a few years time.
/point, laugh - can we get some new Sunshienrs, please, as the current ones are just boringly repetitive and hardly offer any sport.
"There seems to be this autonomic belief that shifting development to India somehow implies a drop in quality and the end to development due to some unspoken belief that Indians can't develop code, just imitate and support Western coders."
That's not entirely true. I worked with a team of "CMM Level 5"-certified coders on a project here in the States -- and they sucked here too!
Interestingly, I think it was the management style and culture that did them in. Individually, each was hard working and no less skilled than most Western coders I've worked with. However, the project was very hierarchically managed, so when a manager or team lead made a bad decision, no one questioned it and it was followed to the letter by all the subordinates. Outside of that culture, I haven't really noticed any skill difference between "local" coders and those from the Sub-Continent.
"OpenVMS [...] like all proprietary systems, [is] under great pressure from the large Windows installed base." Windows, as your recent article on WGA/WAT reminded us, is of course completely non-proprietary.
HP/Compaq, what a bunch of DEC-heads.
The infamy of the Don continues in the squander of the remains of once good company.
OpenVMS will be as ass raped as Vista is now.
Whenever these things are done the old chestnut "You get what you pay for" comes to mind.
We put Bill here because M$ has been doing the same thing.
"Microsoft decommitted from Itanium...."
Not really. The Itanium version is available for Server 2008 R2.
Mr Bryant seems more then a little sensitive to any suggestions that things are not all they could be in the land of HP-UX and Itanium. They most amusing thing is that he automatically assumes any criticism of a HP product must be from a Sun fanboy. I think most people especially those at Intel know that the Itanium has not been and never will be the success it was hoped to be.
Another nail in the Itanium coffin. Sparc, Itanium and just about every other architecture are on a deathmarch organised by the Intel's favoured child the x86. x86 is the turkey of computing and unfortunately its hard to soar like an eagle when you're surrounded by turkeys.
Paris because eveyone should shed a tear for the victims of the x86.
RE: Bit sensitive are we Matt?
"Mr Bryant seems more then a little sensitive to any suggestions that things are not all they could be in the land of HP-UX and Itanium....." Simply tired of the Sunshiner male bovine manure most of us have had to suffer for years, and really happy at the prospect that the source has just about withered and died. It will be interesting to see how much longer the Sunshiners can keep up their fantasies. Why should we let the Sunshiners post more junk when they have finally been exposed for the complete frauds they are? I am assuming the recent zombification of SPARC will mean we will now see an upsurge in Power fanbois posts, but their posts do seem a bit more lucid than the Sunshiners. In the meantime, please point to the RISC architecture you think is doing better than Itanium, which is making at least Intel and hp profits.
RE: Itanium also on Sparc deathmarch
<Yawn> And any facts to support your "assessment"? The only turkeys seem to be the Sunshiners still gobbling away about SPARC, Niagara, etc, etc. The simple fact is the market chooses and it hasn't chosen Sun for a long time. If you quit being bitter and techno-snobbish you'd realise two things; Intel's x86 architecture is the market-leader for a reason - more people bought it than SPARC, Power, MIPS or anything else; and that Intel won't kill Itanium whilst it keeps pressure on Power. After all, the strategy of squeezing SPARC from above with Itanium and below with Xeon is what killed Sun.
"I think most people especially those at Intel know that the Itanium has not been and never will be the success it was hoped to be."
Wasn't Itanium going to be the next big thing? Intel imagined x86 would stay on the desktop and Itanium would rule the datacenter. Didn't Dell and a bunch of others initially line up to flog it then dropped it because the performance sucked?
Points victory because Mr Bryant doesn't know when he's taken a pretty obvious flamebait.
"....Wasn't Itanium going to be the next big thing?...." Do you really want me to post some of McNeedy's or Ponytail's hilarious projections for Niagara, Slowaris 10 or Rock take-up? Wasn't Rock supposed to be running everything on Wall Street by now? Let's see - SPARC is zombified because Sun is dead and about to be butchered up by Larry; Itanium is alive and well and gaining marketshare, especially in hp's Integrity. Intel and hp are posting more profits, whilst Sun has just more and more and more losses, and has been bought for chump change. Forget marketing predictions, just try and admit which option has actually worked out better. Go on, be honest for once, just try. Make sure you're sitting down when you do as your fantasy world is liable to collapse.
"....Points victory because Mr Bryant doesn't know when he's taken a pretty obvious flamebait." Mr Sunshiner would like to call a victory from the deck of his sinking ship? Please, go play with the other children, leave the IT to the adults. All your silly posts do are offer more opportunities to expose yours and Sun's failings.
/Gonna need an arm rest for all this pointing and laughing.
Matt, just stop it. You're making yourself look like a silly fanboi.
We're seeing an uptake in pseries. Sun has really frightened their base. We have a POC
with p6 and PowerVM, when successful will replace 70 Sun boxes down to 4 or 5 , 4U
pseries boxes. The Sun admins admit that LDOMs suck in comparison to pseries LPARs.
Go out and look at Sun quarterly numbers, they are fleeing for the exits folks. It will be fun
to watch that smarmy Larry Ellison deal with his new Sun boat anchor. Good luck with
Biting the hand that feeds IT © 1998–2017