Beta
×

Welcome to the Slashdot Beta site -- learn more here. Use the link in the footer or click here to return to the Classic version of Slashdot.

Thank you!

Before you choose to head back to the Classic look of the site, we'd appreciate it if you share your thoughts on the Beta; your feedback is what drives our ongoing development.

Beta is different and we value you taking the time to try it out. Please take a look at the changes we've made in Beta and  learn more about it. Thanks for reading, and for making the site better!

Microsoft Invests In Open Source Software Company

timothy posted more than 3 years ago | from the world's-biggest-open-source-vendor dept.

Microsoft 99

joabj writes "In what may be its first investment in an open source software company, Microsoft has quietly invested in TurboHercules, which maintains the Hercules open source IBM mainframe emulator. Perhaps the potential for purloining customers from the juicy mainframe market outstrips any misgivings Microsoft may have about open source. You might remember TurboHercules: In March, it filed an antitrust complaint with the EU over IBM's tying of its mainframe OSes with its hardware." A story from earlier this year gives more information on the related conflict between Hercules and IBM over patents.

cancel ×

99 comments

Sorry! There are no comments related to the filter you selected.

misgivings? (2)

clarkn0va (807617) | more than 3 years ago | (#34422548)

Perhaps the potential for purloining customers from the juicy mainframe market outstrips any misgivings Microsoft may have about open source.

The only misgivings MS ever had about open source is for the potential it has for giving away what it has always charged money for, thus eroding their profit share. I've often wondered why they don't leverage it to their own advantage more, much like the way they appropriated BSD code for much of their networking utilities, like netstat et al.

Some FYI (entire MS IP Stack is BSD based)... apk (-1)

Anonymous Coward | more than 3 years ago | (#34423218)

"much like the way they appropriated BSD code for much of their networking utilities, like netstat et al." - by clarkn0va (807617) on Thursday December 02, @03:44PM (#34422548) Homepage

Heh, some "FYI" 4U: Microsoft's entire IP stack is based off the BSD model... & there's even discussions with proof of it online, no less, if you look!

APK

Some actual proof of MS IP Stack being BSD based (-1)

Anonymous Coward | more than 3 years ago | (#34423358)

http://lkml.indiana.edu/hypermail/linux/kernel/9906.2/1214.html [indiana.edu] [indiana.edu]

There's one such example of proof of what I stated here -> http://yro.slashdot.org/comments.pl?sid=1893378&cid=34423218 [slashdot.org] [slashdot.org], & if you search it on GOOGLE ("Microsoft IP Stack" and "BSD") via this query:

http://www.google.com/search?hl=en&source=hp&q=%22Microsoft+IP+Stack%22+and+%22BSD%22&btnG=Google+Search [google.com] [google.com]

You get even more...

APK

Re:Some actual proof of MS IP Stack being BSD base (1)

Anonymous Coward | more than 3 years ago | (#34424470)

"Disregard" that: "I" suck "cock" 4U!

APK

Trolls: Ordinarily I don't IMPERSONATOR trolls but (0)

Anonymous Coward | more than 3 years ago | (#34425414)

""Disregard" that: "I" suck "cock" 4U!

APK" - by Anonymous Coward on Thursday December 02, @05:44PM (#34424470)

Per my subject-line above: Ordinarily I don't feed trolls, even ones that like to impersonate me (5th time in 3 days now), & it's amusing to watch the fool doing so waste his time on it. It's not fooling anyone & it's not like I don't see it!

Keep wasting your time, it's your life.

APK

P.S.=> LMAO - & my posts were "modded down", doubtless by this effete impersonator of myself in this troll (via his doubtless numerous registered user accounts here), but I don't give a hoot about "mod points" or "karma" that much anyhow, OR, I'd be a "registered user" here (and you can clearly see, that I am not)... apk

Re:Trolls: Ordinarily I don't IMPERSONATOR trolls (1)

Inner_Child (946194) | more than 3 years ago | (#34426988)

While you sometimes have valid points, your style is quite... unique, and that's made it prone to mockery, unfortunately. An account wouldn't be the worst idea, even if you don't care about mod points or any of the rest of it. After all, signing your AC posts kind of defeats the point of posting as AC.

Re:Trolls: Ordinarily I don't IMPERSONATOR trolls (0)

Anonymous Coward | more than 3 years ago | (#34429368)

I don't need an account its not like anyone could fake being me.

APK

Re:Some actual proof of MS IP Stack being BSD base (1)

Bigjeff5 (1143585) | more than 3 years ago | (#34426400)

You realize your only proof is speculation that is 10 years and three full OS versions old, right?

That's about as weak as you can get.

Re:Some FYI (entire MS IP Stack is BSD based)... a (3, Insightful)

shutdown -p now (807394) | more than 3 years ago | (#34424564)

Microsoft's entire IP stack is based off the BSD model

It's a very old factoid that became an enduring myth a long time ago. It was really only true back in the days of Windows NT 3.1, the TCP/IP stack for which was a third-party implementation bought by MS. That one was mostly BSD-derived. Since then, however, it was rewritten from scratch (several times, in fact), and NT 3.5 and 95 already included that rewritten version, which is not derived from BSD.

However, the original userland utilities (nslookup, ftp, telnet, a bunch of other stuff) were originally BSD-derived and remain such. That's where the strings "Berkeley" etc (which are usually used as a proof of BSD derivation) come from. So GP is absolutely correct.

Here [kuro5hin.org] is a more detailed treatment of this.

He's correct, I never said he wasn't (I was also) (1)

Anonymous Coward | more than 3 years ago | (#34425514)

"It's a very old factoid that became an enduring myth a long time ago. It was really only true back in the days of Windows NT 3.1, the TCP/IP stack for which was a third-party implementation bought by MS. That one was mostly BSD-derived." - by shutdown -p now (807394) on Thursday December 02, @05:50PM (#34424564)

That's where I was "coming from", but some of the sources in my search query also said it was the same for Windows 2000 (in beta, which IS possible).

I say it was possible, especially during beta for Win2k, because Windows 2000 was the "Direct Descendant" of Windows NT (of which you note now, good post by the by, if I forget to say it, on YOUR part).

I know that the IP stack altered, substantially, in Windows 2000 (when TCP/IP took over vs. LanManager/NetBIOS based communication being the default before that in NT 4.x downwards as one major change as an example thereof), so you're probably correct that it's now "Microsoft Code" via a rewrite... but from myself having written millions of lines of code in my time, I can tell you right now, that I severely SEVERELY doubt it's a "COMPLETE REWRITE" because it works the same for the MOST part (boot up turn on of the IP stack changed since 2000 too, it kicks on FULLY once a user or the system makes an internet bound/IP bound call from Windows XP onwards, as an example for that much also)...

I.E.-> You might be surprised how little is truly "complete rewrites" out there that says it is, & mainly because most coders use prebuilt routines they have stashed over time, that came from WHO KNOWS WHERE, usually code from books adapted has been something!

Thanks for the link too - I can use it in the future during posts like this one I suppose, as a reference.

He's correct, the GP, as you said... I never said he wasn't, but I did want to point out that the IP Stack had BSD roots is all (historical trivia type thing).

APK

P.S.=> Nice post shutdown -p now - if I had "mod points"? I'd give you an upwards mod, but alas, I do not (I don't keep a registered user account here)... apk

Re:Some FYI (entire MS IP Stack is BSD based)... a (1)

Bigjeff5 (1143585) | more than 3 years ago | (#34426644)

I found that thread enlightening.

MICROSOFT FUNDING IBM LAWSUIT (0)

VGPowerlord (621254) | more than 3 years ago | (#34422554)

Cue "Microsoft is funding TurboHercules lawsuit against IBM" conspiracy theorists in 5... 4... 3...

Re:MICROSOFT FUNDING IBM LAWSUIT (0)

Anonymous Coward | more than 3 years ago | (#34422654)

4...5...6... (shh, they might fall for it)

Re:MICROSOFT FUNDING IBM LAWSUIT (0)

Anonymous Coward | more than 3 years ago | (#34422740)

It's not a conspiracy theory if it's true.

Re:MICROSOFT FUNDING IBM LAWSUIT (2)

KillAllNazis (1904010) | more than 3 years ago | (#34424450)

Why can't it be a conspiracy theory and be true?

Re:MICROSOFT FUNDING IBM LAWSUIT (0)

Anonymous Coward | more than 3 years ago | (#34424578)

if you know it's true, it's no longer a theory. eg: relativity, evolution, gravity, etc, are theories -- not proven facts. Some people might think that man descended from monkeys or the osama bin laden was responsible for 911 or Microsoft is funding the TurboHercules lawsuit against IBM, but there is no proof.

Re:MICROSOFT FUNDING IBM LAWSUIT (5, Insightful)

killmenow (184444) | more than 3 years ago | (#34422862)

The fact is Microsoft is funding TurboHercules and thereby funding the lawsuit. Now why Microsoft is funding TurboHercules may have little or nothing to do with said lawsuit.There's room for conspiracy theories there and those who are into such things can (and likely will) take that ball and run with it but I don't know or care to speculate. Still, "Microsoft is funding TurboHercules lawsuit against IBM" [sic] is a statement of fact. They weren't funding it at the start, but they are now.

Re:MICROSOFT FUNDING IBM LAWSUIT (1)

harlows_monkeys (106428) | more than 3 years ago | (#34423522)

To the conspiracy theorists (hello, Boycott Novell!), Microsoft was secretly behind the lawsuit from the start, and this is seen as some sort of proof they were right about that.

Re:MICROSOFT FUNDING IBM LAWSUIT (0)

PPalmgren (1009823) | more than 3 years ago | (#34422912)

You know one thing I love about conspiracy theorists? They really make me appreciate my sanity. Watching those wackos on the history channel conspiracy shows gives me a huge appreciation for rational thought.

Re:MICROSOFT FUNDING IBM LAWSUIT (1)

nabsltd (1313397) | more than 3 years ago | (#34423540)

Watching those wackos on the history channel conspiracy shows gives me a huge appreciation for rational thought.

Admit it, "they" have gotten to you. What did "they" have to do to ensure your silence?

Re:MICROSOFT FUNDING IBM LAWSUIT (1)

tehcyder (746570) | more than 3 years ago | (#34429482)

I hope you'll be happy when your alien overlord paymasters are eating your brains.

Re:MICROSOFT FUNDING IBM LAWSUIT (2, Interesting)

Anonymous Coward | more than 3 years ago | (#34423022)

MSFT have form, remember SCO? Why else would they invest in this company?

Re:MICROSOFT FUNDING IBM LAWSUIT (1)

poetmatt (793785) | more than 3 years ago | (#34423046)

really? It's pretty much actual confirmation of it.

Yea that is interesting... (2)

LWATCDR (28044) | more than 3 years ago | (#34422564)

If the courts make IBM give in this could be huge for Microsoft. Turbo Hercules could be used to make an easy migration to Windows. Or maybe Microsoft will make an IBM mainframe compatibility layer like the Posix layar using Turbo Hercules? If you could run your mainframe software on a Windows server things would really start to suck for IBM.

Re:Yea that is interesting... (2)

MBGMorden (803437) | more than 3 years ago | (#34422708)

Indeed. Not to mention it might actually be feasible for some people to learn mainframe OS's in their spare time.

We've got an AS/400 at work that we still do a lot on (though we've been migrating away from it slowly). It's the only system I don't have a fucking clue how to use. I mean sure, I can login, and futz around enough to get a few basic things done, but I don't really understand the system, and that bugs me.

Windows was easy. For $200-300 you can get a Windows system. Linux? About the same or less. Mac? Well, with some creative VMWare or Hackintosh tricks you can run it on regular on PC hardware, or if you need to buy one the Mini is cheap. Of course I personally learned OS X on a used G4 tower I grabbed for $99.

Point is, a hobbyist can learn this stuff at home. Mainframe stuff though? Good-fricken-luck. That hardware and software is priced so far out of reach of the consumer that almost no one is going to lay down the cash for one - paritcularly given their limited functionality in regards to things a home user would find interesting.

I don't care how bad the performance was - I'd kill for a nice AS/400 emulator just so that I could learn the ins and outs of it. Break it, reinstall, and go again. The same way I learned every other OS. Alas, I have no intention of just "playing around" on my employer's production system, so I remain largely in the dark.

Re:Learn Mainframe OS (1)

TaoPhoenix (980487) | more than 3 years ago | (#34422812)

Is there anyone online that lets you remote in to a public copy dedicated for educational purposes?

Re:Learn Mainframe OS (1)

MBGMorden (803437) | more than 3 years ago | (#34423042)

You can get accounts on them for fairly reasonable prices, but you're essentially buying time as a user. From what I've seen it's more useful for developers who want a machine to code on when they're aware from work. From a system administrator perspective, you just don't have the access capabilities to learn how to do things like OS upgrades and the like.

For Unix comparison, it'd be like being given an account on a machine with no root access. You can code, run apps, etc, but no fiddling with the setup.

Re:Learn Mainframe OS (0)

Anonymous Coward | more than 3 years ago | (#34423568)

IBM hold a student Mainframe contest every year. I've taken part for the last three years and it's not too bad. I can never be bothered to finish the second task though. I don't really like how structured the environment is. If I could just go play with a mainframe from the word go it would be much more interesting; install, break, install, break... Which is how I learned UNIX, the Mac (twice), Linux and obviously Windows.

Re:Yea that is interesting... (2)

baegucb (18706) | more than 3 years ago | (#34425758)

An AS/400 isn't a mainframe. It's based on a System/38, a mini-computer, iirc (I've used both but not recently). Anyways, the Hercules emulator is based on MVS, which was originally open source from IBM. MVS begat z/OS which is the current IBM mainframe OS (we also run Linux LPARs using virtualization and specialized processors). And no, x86 doesn't have the IO capability to compete with mainframe hardware. The most Hercules could do is try and convinve mainframe customers to run their software, rather than something like z/OS. And I don't see that being done say by a bank or credit card company, for instance.

Re:Yea that is interesting... (1)

bws111 (1216812) | more than 3 years ago | (#34425934)

Hercules is not based on MVS in any way. Hercules is based on the z/Architecture Principles of Operation, which describes the instruction set, etc for the z/Architecture. Hercules (+windows+pc) is a 'replacement' for mainframe hardware, nothing else.

Re:Yea that is interesting... (2)

baegucb (18706) | more than 3 years ago | (#34426250)

Ummm.. nope. Hercules initial release was 1999 (according to wiki although I was relying on memory in my first post). z/OS came out in 2001. It may now use Z architecture, but not originally.

Re:Yea that is interesting... (1)

bws111 (1216812) | more than 3 years ago | (#34426598)

OK, so originally it was based on the ESA architecture, now it is the z/Architecture. Still never had anything to do with MVS at any point in it's life.

Re:Yea that is interesting... (1)

baegucb (18706) | more than 3 years ago | (#34426662)

That was the main purpose back then. To run MVS on a PC. http://www.jaymoseley.com/hercules/ [jaymoseley.com] for example.

Re:Yea that is interesting... (2)

mswhippingboy (754599) | more than 3 years ago | (#34425946)

Hercules emulates the IBM large systems machines (360/370/390/ES9000/zSeries), no AS/400 support here. The AS/400 was (is) a midsize machine (we used to call them minicomputers back in the day). It is a totally different beast from the large systems. If I recall, when it was introduced it had a lot of unique features (variable length words, memory mapped files), at least for it's time. Later machines switched to the Power architecture (iSeries I believe).
However, I never cared for it because (as a developer) I always felt boxed in because everything was done through dialog panels and I never felt I could get close enough to what was going on inside the machine - and RPG III sucked!.
Someone used to sell a software/hardware version of the AS/400 (baby/400 I believe it was called), but I'm sure it's long gone - it was pretty expensive when it was around anyway.

Re:Yea that is interesting... (1)

dwywit (1109409) | more than 3 years ago | (#34427160)

*Sigh* - let me know when you find as AS/400 emulator - I'd like to see how to make it work properly on non-AS/400 hardware, but just having that OS to play with.... I've been out of it for a while, but I've never seen the granularity of control available in any other mid-range or server OS. I've never seen a Microsoft OS that even approached the ability of OS/400 to manage resources and jobs. FWIW, I used to manage an AS400 that supported 200+ green screens and 200+ desktop PCs. It had 48MB (yes, megabytes) of main memory and still gave sub-second response times on green-screen applications. The PCs only used it for storage and printing. BTW, http://systeminetwork.com/ [systeminetwork.com] is a good place to start if you want to learn more.

Re:Yea that is interesting... (1)

tehcyder (746570) | more than 3 years ago | (#34429498)

I have no intention of just "playing around" on my employer's production system

Wuss! What's the worst that could happen?

Ah...

Re:Yea that is interesting... (1)

Anonymous Coward | more than 3 years ago | (#34422732)

Meh, I don't know about that. I don't think we would see people flocking from mainframe hardware to a Herc/Win box for production stuff. Hercules is very cool, but if you poke around with the software you get the feeling that it's not something you would bet your F500 company on.

Re:Yea that is interesting... (1)

bhcompy (1877290) | more than 3 years ago | (#34423196)

It's a start. PICK OS is currently emulated very successfully, with APIs built in to the emulator for TCP/IP communication(PICK predates TCP/IP), SQL interaction(PICK is a multivalue db with its own query language, not a relational db with a dialect of SQL), etc. You don't buy mainframes to run PICK anymore, you emulate it in *nix or Windows. When the Hercules software matures, the same thing will probably start to happen because of the cost and maintenance.

Re:Yea that is interesting... (4, Interesting)

bws111 (1216812) | more than 3 years ago | (#34423580)

Hercules is a hardware emulator. Running stuff on Hercules does not get you any closer to migrating to Windows than running on a real zSeries machine does. The only thing Hercules does is allow you to move from expensive but highly reliable hardware to cheap hardware. Of course that move comes with an enormous performance penalty, and your 5 9s mainframe reliability has just gone in the toilet. There are probably only a handful of IBM customers world-wide who would even seriously consider doing that. The only thing Microsoft 'gains' from this is potential damage to IBM.

Re:Yea that is interesting... (1)

inode_buddha (576844) | more than 3 years ago | (#34424416)

+3, insightful (if I had any points...)

Re:Yea that is interesting... (1)

larry bagina (561269) | more than 3 years ago | (#34424658)

They gain the employees. On a related note, Microsoft also bought up VirtualPC a few years ago.

Re:Yea that is interesting... (1)

LWATCDR (28044) | more than 3 years ago | (#34426186)

Actually you will not loose too much if any performance depending on which zSeries you are using. Of course Microsoft would have to write a compatibility layer on top of the hardware emulator for a complete solution.

Re:Yea that is interesting... (2)

baegucb (18706) | more than 3 years ago | (#34426326)

You lose reliability and speed of IO amongst other things. Start at about page 11 of http://www.redbooks.ibm.com/redbooks/pdfs/sg246366.pdf [ibm.com]

Re:Yea that is interesting... (1)

LWATCDR (28044) | more than 3 years ago | (#34430976)

Reliability is a given. There are few things in this life that are as Reliable as a well maintained Z Series. But Intel based servers are getting very good at up time these days. Redundant power supplies and RAID arrays go a long way. Don't get me wrong. They are not as reliable as an IBM mainframe. Heck you can hot swap CPUS on those things but they are pretty good.
Herc on a PC could replace some smaller mainframes. A company might decided that the uptime of their windows servers is now good enough for their needs so they can migrate away from mainframes.
Plus as more mainframe style apps are moved to Wintel you will companies offer more mainframe style hardware features for the platform. Some of the new serves have some pretty advanced management features now.

Re:Yea that is interesting... (1)

Just Some Guy (3352) | more than 3 years ago | (#34430836)

There are probably only a handful of IBM customers world-wide who would even seriously consider doing that.

I knew a friend who did some work for a ma-and-pop trucking company to run the AS/400 they were tricked into buying. They have about 5 trucks on the road in peak season. How ultra-reliable does their system need to be? Sure, you don't want it to lose or corrupt data, but if push came to shove they could leave it turned off 23 hours a day and only boot it for some occasional data entry.

I'd stake money that there are a lot of little shops like this with way more computing hardware than they could possibly need. For someone like this, an emulator might be a perfectly reasonable step in migrating from an expensive overkill system to something more appropriate.

Re:Yea that is interesting... (1)

bws111 (1216812) | more than 3 years ago | (#34431952)

That may be true, but it has nothing to do with Hercules or this story. AS/400 is what used to be called a mini-computer. It was sold as a turnkey system to small and medium sized businesses, and as a departmental server in large businesses. It has it's own processor architecture and instruction set, and it's own operating system (OS/400). Hercules, on the other hand, is emulating the z/Architecture machines. These machines have a completely different architecture (which started out as S/360) and operating system (z/OS, formerly MVS). zSeries machines have price tags in the millions and are purchased mostly by Fortune 100 type companies. The software licensing costs can be in the 10's of thousands of dollars each month.

So, the customer set for a product like this would be zSeries customers who don't need the performance or reliability of a mainframe, but are still willing to pay IBM thousands of dollars in license fees each month rather than port their application to a native *nix or Windows solution. That seems to be a very small set.

Re:Yea that is interesting... (1)

LWATCDR (28044) | more than 3 years ago | (#34433334)

There is a "low" end System Z both for the 9 and 10. According to IBM they start at around $100,000. while not cheap is a lot less than millions of dollars.
And let's not forget things like development, testing, and even legacy systems that people may be sill running. I am sure that there is more 370 and 390 code still running that any of us wants to think about.

Re:Yea that is interesting... (1)

rtb61 (674572) | more than 3 years ago | (#34432536)

It doesn't seem likely that the intent is to damage IBM so much as a opportunity to build up a series of legal cases against IBM to be used in negotiation with IBM over something like Linux patents. So IBM produces an OS that they sell with their hardware and they want to control licensing of that OS ie if you run it on other hardware you must be pirating it as they only sell it with their hardware, this case nor the other one doesn't really seem to have any depth.

So M$ just seems to be temporarily propping up those companies whilst the spend money on legal challenges against IBM, to put pressure on IBM. The current biggest OS threat to windows is of course Android and M$ seems to be playing the game of circling their legal wagons to build up a monopolist defence.

Re:Yea that is interesting... (1)

bws111 (1216812) | more than 3 years ago | (#34432750)

After reading some of the other posts I thought of another reason for MS to do this. For example, there is a post by someone whose company has a mainframe, but it is very lightly used and the rest of the stuff is Microsoft. It seems to me that that lightly used mainframe represents a big threat to MS, in that someone could wake up one day and say 'hey, we could move a couple hundred web servers and SQL servers to run on that mainframe with Linux', and there goes MS's grip on that customer. So it is advantageous for MS to push for a product that would remove the lightly used mainframe before that can happen.

Re:Yea that is interesting... (1)

LWATCDR (28044) | more than 3 years ago | (#34423556)

What if it is a Microsoft supported and endorsed Herc/Win box?
I think you may be under estimating how much trust people have in Microsoft.
And people are already using Herc for things like development and yes even for replacing older mainframes.

Re:Yea that is interesting... (1)

TheRaven64 (641858) | more than 3 years ago | (#34424770)

Depends on why they're using a mainframe. If it's because a mainframe is the right tool for the job, then Windows almost certainly isn't. A lot of people, however, are using a mainframe because it was the only thing that could handle the workload back in the '70s or '80s. These days, a cheap desktop can probably handle the load if it hasn't grown much.

There are also the people developing mainframe software. They could easily work on a laptop and only deploy on the real hardware. Moving these people off the mainframe reduces the load on the mainframe, taking money from IBM (they charge for turning on each CPU).

Re:Yea that is interesting... (1)

evilviper (135110) | more than 3 years ago | (#34425252)

They could easily work on a laptop and only deploy on the real hardware. Moving these people off the mainframe reduces the load on the mainframe, taking money from IBM (they charge for turning on each CPU).

Except it practically never works that way... making it cheaper and infintely easier for people to develop and test software for IBM mainframes is likely to make the platform much more popular, not less. And realistically, would you expect companies to run critical software on an emulator? Personally, I think this will, at best, eat the low end of IBMs market... those who want to run a handful of legacy apps without the money or time to rewrite them. They aren't cash cows, often buying old used equipment, keeping it running for decades, and if anything, forcing IBM support to keep familar with ancient platforms and all possible compatibility issues that might arise.

Re:Yea that is interesting... (2)

bws111 (1216812) | more than 3 years ago | (#34425548)

Correct. IBM itself has offerings to allow you to emulate a zSeries on a linux box for development purposes. [ibm.com]

However, I must point out that the EU complaint is not about IBM objecting to companies that are running old stuff using Hercules. IBM is objecting to companies running the very latest z/OS on an emulation of IBM's latest hardware.

Re:Yea that is interesting... (1)

Ritchie70 (860516) | more than 3 years ago | (#34427076)

I think my (Fortune 200) employer has a bunch of stuff that runs on a mainframe because that's where it runs, not because of any volume or even reliability requirements. It needs to run once a day (or once a week) and process what once was an awe-inspiring amount of data but now is a DVD.

Since we're officially a "Microsoft Shop" I could totally see us using something like this.

I am kind of amused by Microsoft now being in the business of "hardware" to run IBM's "software" though.

Wow! Do you honestly think... (0)

Anonymous Coward | more than 3 years ago | (#34425640)

...that companies that can afford to run mainframe systems would still be running some 1970/80's app that could have been migrated to a low-end platform would still be running it on a mainframe?

Your naievety and ignorance is astonishing.

And FYI development load on a z/OS mainframe is a flea bite on an elephant. We're not talking about crappy *nix boxen here.

Re:Yea that is interesting... (1)

mswhippingboy (754599) | more than 3 years ago | (#34426034)

Many F500 companies have legacy mainframe applications that are not critical to the business, but they don't want to spend the $$$ to rewrite them for another platform. For these application, Hercules would make sense (if they could legally run z/OS). Also, whenever I'm engaged in a mainframe development project (yes, they still happen - occasionally - though usually only for maintenance enhancements), I use my local MVS instance to do the development since the "real" mainframe is usually pegged and just getting a compile/link/go through can be frustratingly slow. In my local instance it flies. Since I'm running on a 3.2Ghz quad-core machine, it's acually far, far faster than the "real" mainframe I used to work on.

Re:Yea that is interesting... (2)

mswhippingboy (754599) | more than 3 years ago | (#34425668)

If you could run your mainframe software on a Windows server things would really start to suck for IBM.

Depending on what you are doing on the mainframe, you can do that already. If all you need is to be able to run some mainframe batch jobs or TSO scripts execs, then it's already there now. Simply download Hercules and you can download your choice of MVS 3.8j, DOS Release 34, or VM/370 (maybe even a couple more). These are quite old versions of mainframe operating systems which are in the public domain. I have all of them installed and occasionally play when them just to keep my old mainframe skills fresh. Unfortunately many mainframe systems use products such as IMS or CICS which are not in the public domain (at least not at this time).
The Hercules emulator can actually even run the current version of z/OS (basically, the latest incarnation of MVS), just not legally, which is what the whole problem IBM has with TurboHercules. TurboHercules wants to be able to legally run z/OS on the emulator for purposes of disaster recovery. IBM says no, z/OS can only be run on IBM hardware.

Re:Yea that is interesting... (1)

bws111 (1216812) | more than 3 years ago | (#34425836)

Has IBM actually said z/OS can only run on IBM hardware? The only statement related to that that I have seen is where they said they wouldn't license for Hercules, because they think Hercules infringes on patents and other IP.

Re:Yea that is interesting... (3, Interesting)

mswhippingboy (754599) | more than 3 years ago | (#34426224)

z/OS is licensed per CPU (tied to the CPU serial number). Long before TurboHercules, the Hercules community tried to get IBM to provide a "hobbyist/student" license to allow z/OS (or OS/390) to be legally run on Hercules, but IBM was not interested. I don't believe at that time IBM was concerned with IP or patent issues, there just wasn't enough in it for IBM to waste their time with.
Having said that, I know IBM "used" to license their OS to run on competitor mainframes such as Amdahl and Fujitsu, mostly because they were forced to as part of the antitrust settlement. I don't know it that's the case anymore.
However, even if IBM were to be forced to license z/OS to be run on Hercules, I'm sure no one could afford it (at least for hobby/training purposes). I remember back 20 years ago we used to pay about $6,000 per month just for MVS (about $25,000 per month for all the IBM software we had), and we were a pretty small mainframe shop.

Does competition law trump patents? (4, Insightful)

ciaran_o_riordan (662132) | more than 3 years ago | (#34422652)

Besides the people behind this case, the case itself is quite interesting too.

The European Commission (or Court of Justice) will have to decide if IBM has harmed TurboHercules through anti-competitive behaviour. IBM has also asserted patents. This means that if the European institutions find that IBM is doing wrong, then they will also have to decide if IBM can use its patents to continue the wrong. I.e. what trumps? Competition law or patents?

http://en.swpat.org/wiki/IBM_and_TurboHercules,_2010 [swpat.org]

http://en.swpat.org/wiki/Competition_law_defence [swpat.org]

If competition law trumps, then this opens a new path for breaking down the problems that software patents are doing to standards and interop.

http://en.swpat.org/wiki/Harm_to_standards_and_compatibility [swpat.org]

Re:Does competition law trump patents? (-1, Offtopic)

commodore64_love (1445365) | more than 3 years ago | (#34422908)

+1 informative.

Re:Does competition law trump patents? (2)

oh_my_080980980 (773867) | more than 3 years ago | (#34423278)

You know this has been decided before in favor of IBM.

If you bothered to research the case you will find the issue is with a company called Turbo Hercules telling companies that have mainframe contracts with IBM to violate those contracts so they can use Turbo Hercules.

It's a matter of contract law.

IBM does not have an issue with the Turbo Hercules project and have donated code to that effort.

IBM does have an issue with a piss-ant company trying to get their customers to violate terms of their agreements.

Moron.

Got a link? (1)

ciaran_o_riordan (662132) | more than 3 years ago | (#34424230)

> You know this has been decided before in favor of IBM.

Oh? Got a link?

My understanding was, as is mentioned in the story summary:

"You might remember TurboHercules: In March, it filed an antitrust complaint with the EU over IBM's tying of its mainframe OSes with its hardware."

I.e. no decision has been taken.

If you can give a link to show it's been decided, I'll be happy to admit I wasn't up to date on the case.

Re:Got a link? (1)

TheRaven64 (641858) | more than 3 years ago | (#34424818)

The previous case was in the 80s and was a US antitrust case. An EU case will be testing different laws and the situation with regard to IBM's monopoly is quite different (people do get fired for buying IBM now), so the outcome may well be different.

This case is particularly interesting because it may set a precedent that prevents tying OS X with Macs, and so on.

Re:Got a link? (1)

ciaran_o_riordan (662132) | more than 3 years ago | (#34425010)

Ok, thanks.

Did the US case in the 80s touch the patents issue at all?

If it was over software, it probably didn't since no one was really using software patents back then, but if there were any patent issues, I'd still be interested in getting a reference so I can look into how the competition authority handled it.

(To restate the broader question: when the competition authority finds anti-competitive behaviour that's being propped up by contract or market forces, and takes the necessary action to end this anti-competitive behaviour, does this prevent the losing party from using their patents to continue to exclude competitors?)

Re:Got a link? (1)

bws111 (1216812) | more than 3 years ago | (#34425132)

Well, in the 50's (US), IBM was forced to license it's patents to anyone who wanted them. If IBM and the other party could not come to agreement on the royalty the court would decide it.

Re:Got a link? (1)

ciaran_o_riordan (662132) | more than 3 years ago | (#34425552)

Hmm, that mention of royalties is usually a bad sign, but it would be interesting to see what might happen if IBM's only competitor was a free software project.

Hate to be a broken record, but have you a link (maybe a Wikipedia page) for that case?

(The reason I keep asking for links is that I'm building a wiki at en.swpat.org which is an information resource for campaigns against software patents. I'm trying to raise the general quality of lobbying, so I try to avoid adding unsourced stories.)

Re:Got a link? (2)

inode_buddha (576844) | more than 3 years ago | (#34426414)

It's a matter of historical fact that IBM was sued twice for antitrust in the US. I don't know the exact case numbers. The first case began in the late 1950's, and the second one was in the 1970's/80's. I recommend you get a copy of "Father Son and Company: my life and times at IBM" by Thomas J. Watson (the son of the founder). You should be able to find it on Amazon, it's a good read of IBM's corporate history as viewed from the inside by the president/CEO.

FWIW, I'm on IBM's side in this matter with TurboHercules, since I tend to read the details of court cases.

Re:Got a link? (1)

ciaran_o_riordan (662132) | more than 3 years ago | (#34426784)

I'm not doubting anyone that these cases took place :-) Thing is, I'm building a wiki of information which should be useful for campaigns against software patents. I'm trying to raise the standard of discussion, so I avoid adding information that doesn't come with proof.

I probably won't get time to read that book.

As for being on one side or the other, I'm on neither. My role is that I'm trying to *document* what happened (insofar as it's relevant to patents and software development). I'm not calling for anyone to win or lose. I'm just pointing out what influence this case might have for software patents (which is of course *if* IBM loses, since if IBM wins, this case can't have any influence on software patents).

Re:Does competition law trump patents? (0)

Anonymous Coward | more than 3 years ago | (#34423282)

You know what real anti-competitive behavior is? Allowing a so called 'competitor' to invest literally billions of dollars in something, then taking the results of that investment (without paying a dime) and making a cheap knock-off to take customers away.

In this case, I think patents are doing exactly what they are supposed to do - encourage innovation by allowing you to reap the rewards. If TurboHercules wants to 'compete', let them do so fairly by paying IBM for their investment.

Re:Does competition law trump patents? (1)

oh_my_080980980 (773867) | more than 3 years ago | (#34423336)

Thank you. And how about Turbo Hercules stops trying to get IBM customers to violate their contract.

IBM is no different than Apple (3, Insightful)

kawabago (551139) | more than 3 years ago | (#34422660)

Apple ties OSX to it's own hardware and no one argues that is wrong, although I think it is.

Re:IBM is no different than Apple (1)

Wesley Felter (138342) | more than 3 years ago | (#34425258)

Did you not hear about the Hackintosh movement and Psystar? Plenty of people have complained about OS X being tied to Macs.

Re:IBM is no different than Apple (1)

bill_mcgonigle (4333) | more than 3 years ago | (#34427482)

Apple ties OSX to it's own hardware and no one argues that is wrong, although I think it is.

I think it's unfortunate, but how is it 'wrong'? Their software, their hardware, their choice.

Re:IBM is no different than Apple (1)

kawabago (551139) | more than 3 years ago | (#34427716)

Software is protected by copyright. As an author I can't control where you read my book. Why should Apple control where we run their software? The law can be wrong and I think it is here, fair use needs to be expanded.

Re:IBM is no different than Apple (1)

bill_mcgonigle (4333) | more than 3 years ago | (#34431326)

As an author I can't control where you read my book. Why should Apple control where we run their software?

Would you use threats of violence (government) to tell the author of that book what to write?

Re:IBM is no different than Apple (1)

tehcyder (746570) | more than 3 years ago | (#34429860)

Apple ties OSX to it's own hardware and no one argues that is wrong, although I think it is.

A lot of people would argue it is wrong, but 99% of Apple customers don't care as long as it's cool and works. It is the 1% who should know better (developers, et al) that annoy me. If it was Microsoft doing this, they'd be all over them for anti-competitiveness, lack of openness and so on, but somehow Apple get exempted because they're technically quite good/

Investment in FOSS (2, Insightful)

Anonymous Coward | more than 3 years ago | (#34422672)

Conversely, IBM should invest serious money/time in ReactOS and WINE ... and encourage the liberation of Mono...

Re:Investment in FOSS (1)

mcneely.mike (927221) | more than 3 years ago | (#34424484)

Say it again, won't you Antony?

Invest in FOSS... (1)

Antony T Curtis (89990) | more than 3 years ago | (#34422680)

IBM should invest serious money/time in ReactOS and WINE ... and encourage the liberation of Mono...

Re:Invest in FOSS... (1)

mcneely.mike (927221) | more than 3 years ago | (#34424500)

echo

Well, it's not like they have another motive.. (4, Insightful)

iONiUM (530420) | more than 3 years ago | (#34422692)

While I'm happy Microsoft is investing in open source, I find that their target is fairly suspicious.. what easier way to take on IBM indirectly than to give money to an open source company who is already in conflict with them.

In addition, it's not like Microsoft isn't already trying to embrace open source. You'd be surprised at just how much stuff is released under MS-PL licence. And while that may anger you, as it's their own licence, it's rather free.

Re:Well, it's not like they have another motive.. (1)

Anonymous Coward | more than 3 years ago | (#34423220)

Exactly - this is SCO in a different hat, but with an actual product.

Re:Well, it's not like they have another motive.. (0)

Anonymous Coward | more than 3 years ago | (#34429322)

Stop being dim. Microsoft are not, "...investing in open source" They are investing in something (which happens to be open source) which will help them undermine a company which currently finances many open source projects.

They couldn't care less about open source except that it is an impediment for them being able to leverage their old monopoly.

Don Quixote de Redmond (2)

Un pobre guey (593801) | more than 3 years ago | (#34422714)

As time goes by, most MS senior management will realize that open source is neither good nor bad, but an instrument that can be more useful than closed source under many circumstances.

En un lugar del estado de Washington cuyo nombre quiero olvidar...

Re:Don Quixote de Redmond (1)

trickyD1ck (1313117) | more than 3 years ago | (#34429226)

I wonder if F$F zealots will ever realize that software is just an instrument.

SCO? (0)

Anonymous Coward | more than 3 years ago | (#34422888)

'First ever investment in an open-source company' - really? What about the money they pumped into SCO (for mostly the same reason).

Embrace, Extend (0)

Anonymous Coward | more than 3 years ago | (#34422968)

Already two steps and you all know what the next step is.

Interesting. (0)

Anonymous Coward | more than 3 years ago | (#34423204)

Isn't apple tying their OS to their hardware (via their licensing terms). Is that also illegal in the EU? And yet in the US, it seems that Apple can take companies to court for using their software on hardware that Apple doesn't want you using their software on.

Warren

Re:Interesting. (1)

HTH NE1 (675604) | more than 3 years ago | (#34423458)

A complaint is not a ruling of law, nor is a settlement a ruling of law. Has the EU even ruled yet?

Microsoft already invested in open-source (0)

Anonymous Coward | more than 3 years ago | (#34423344)

Apache is the most notable example. Microsoft fight copyleft (GPL) by supporting projects that use permissive licenses, open-source software distributed under permissive license could not compete with corporations.

Bruce Perence explained this two years ago: http://itmanagement.earthweb.com/osrc/article.php/3762786/Bruce-Perens-Microsoft-and-Apache---Whats-the-Angle.htm [earthweb.com]

Wait, what?! (1)

skywatcher2501 (1608209) | more than 3 years ago | (#34423520)

M$.. investing in open source.. good.. evil.. must.. classify.. DOES NOT COMPUTE!

Re:Wait, what?! (1)

Bigjeff5 (1143585) | more than 3 years ago | (#34427120)

Clearly they are doing good for the purposes of evil!

Think slashdot could get anything straight... (0)

Anonymous Coward | more than 3 years ago | (#34424412)

Like the fact that TurboHercules is no more an open source company than Oracle is.

The open souce project Hercules was started around 1999. TurboHercules was started in 2009.

Last I looked TurboHercules did not sell Hercules but sold systems on which hercules ran on.

So if Dlink used linux in it's routers and Microsoft invested in Dlink would it be investing in a linux company?

I think not.

On top of that, Microsoft is not investing in TurboHercules to help develop Hercules, rather it is investing In TurboHercules to help it fund lawsuots against IBM, the same way Microsoft invested in SCO.

more uninformed garbage (0)

Anonymous Coward | more than 3 years ago | (#34425078)

"In what may be its first investment in an open source software company

HUMM........Microsoft invested heavily in NOVEL.....SuSe LINUX? And Everyone on slashdot picked the sale of Novel apart specifically the details about MS Investment. Now everyone wants to play dumb as if this is NEWS? IDIOTS.

Integration at the UI layer (0)

Anonymous Coward | more than 3 years ago | (#34425496)

My company uses IBM mainframes and trying to get in and out of our databases is a nightmare - they're not relational, they're often CA Datacom (see the recent CA copyright story on Slashdot). There seems to be an increasing amount of companies that are building integration at the UI layer, i.e. they have a product that can recognise various IDs in a terminal emulator and then bring up a side panel on your desktop that links in information related to it. Example: looking at a customer on your mainframe system through a standard terminal emulator on Screen X, the third party program recognises you're on Screen X, picks up the ID and brings up all the customer information from your Siebel DB in a side panel.

Client level integration with your competitors' closed systems: cheaper as long as you have a way of getting into the client of your competitor.

With this acquisition, Microsoft go from Windows integration to mainframe. That gets them into most of the key systems of the Fortune 500, if it's good enough.

Re:Integration at the UI layer (1)

baegucb (18706) | more than 3 years ago | (#34426480)

I know of a great tool to help convert your DATACOM databases to DB2. It's called 2BDB2 at http://www.isi.com.au/2bdb2.html [isi.com.au]

TurboHercules is not an Open Source company (0)

Anonymous Coward | more than 3 years ago | (#34429414)

"TurboHercules SAS (“TurboHercules”) operates the TurboHercules.com Web site. You have the right to use the Web site and download material from this Web site for your personal, noncommercial use. Material may be purchased and downloaded for commercial use, subject to the terms and conditions of the applicable end user license agreement link [turbohercules.com]

"The Q Public License (QPL) is a non-copyleft license, created by Trolltech for its free edition of the Qt. It is incompatible with the GPL, meaning that you cannot legally distribute products derived from both GPL'ed and QPL'ed code link [wikipedia.org]

IBM should take a leaf from the Microsoft litigation manual and threaten to sue TurboHercules end users.

Not really investing in Open Source itself (1)

ShOOf (201960) | more than 3 years ago | (#34430364)

They aren't investing in the company because it's open source, they are investing in a company that just happens to release their product as open source.

Check for New Comments
Slashdot Login

Need an Account?

Forgot your password?