Slashdot Mirror
Interview With IBM's Chief Linux Strategist
Linux Magazine interviewed
IBM's chief
Linux strategist, Irving Wladawksy-Berger, about IBM and Linux. IBM sees Linux as a
disrupting technology of the same class as the Internet: an OS that can run on
many platforms and that nobody owns: something that can fundamentally change
the landscape of computing. By adding Linux compatability to AIX/Monterey, IBM
is guaranteeing itself a big-iron version of Linux without angering the
community by forking the kernel ... but they'd obviously also would like to see
big-iron features added to Linux. Interestingly, Irving suggests IBM would be
willing to open-source just about anything the Linux community wants ... one
just needs to make one's mind known at IBM's
developerworks site. Following the usual path, Linux adoption by
IBM was a bottom-up process, finally convincing senior management.
It's a shame that Linux Magazine did not ask about IBM's patent strategy, which was
already a hot topic two years ago when
slashdot
facilitated the release of the Jikes compiler.
Does that mean that IBM will put these major parts of Linux/S390 under an OpenSource license, as well?
-- jochen
1) Um, IBM is making oodles of money off the internet. A massive portion of the services business comes from "e-business" stuff sold to Fortune 500 companies. (IBM's traditional customers.) Also, DB2 and WebSphere are getting to be pretty popular among enterprise customers.
2) IBM doesn't make any money off of AIX. Yes, they do charge licensing fees, but mainly it is sold as something you need to buy to run your RS/6000.
IBM makes it's big money from the holy trinity of high-end hardware, industrial-strength software applications, and services/solutions. Low-end hardware (like 1U servers and PC's) and commodity software (like OS's) are only provided because they need to be if IBM wants to sell a "solution" instead of merely products.
IBM's other major profit streams are Technology (patent portfolio, Chip fab services, etc.), ThinkPads (simply because they kick ass and can take advantage of IBM's research in displays, hard drives, etc.), and leases/loans so customers can afford all this stuff.
If Linux could do what IBM needs, (more reliability, better I/O, etc.) they would likely drop AIX like a hot potato, because developing it ain't cheap.
Remember that the third letter in IBM is "Machines". IBM has always been more interested in selling hardware than software. If IBM provides top-notch support for Linux across their entire hardware line, it'll attract more buyers of their hardware as more and more businesses use Linux.
Of course, this is soon not a problem anymore, as Linux has been ported (and at least successfully booted) on E10K now. A post on LKML two days ago was on this very subject (and NO, I'm not confusing things with the Alpha announcement, which was made a few hours later; one 20K BogoMIPS (the E10K) and one 40K BogoMIPS (the Alphaserver) computer announced on the kernel-list in just a few hours. That's majorly cool. I wish I had one, and someone who could pay the electricity.
I think it might be related to the remark about benchmarks -- Linux advocates and corp.s (such as Redhat, who funded development the "Tux" webserver) wouldn't want to be shown up by IBM when the results IBM returned showed the original benchmarks to have been overblown.
At least that's how I read that part.
"Oh, I hope he doesn't give us halyatchkies," said Heinrich.
Finish giving us the rest of the open source for AIX's LVM/JFS, virtual memory manager and diskcache buffering sourcecode and we'll be happy for a for a while... then we'll start screaming for DB2's source.
I'll admit that the default color scheme was designed to drain one's life force out through one's nostrils, but once customization of colors and behaviors was engaged in by the knowledgable user, its beauty had no limits.
If only I could go back to the days of reworking the launchpad with a few short sweet lines of rexx. ....
have a day,
-l
OS/2 is still partly copyrighted by Microsoft; rumor has it they cannibalized large amounts of it for NT. Microsoft would probably pull a Chernobyl if IBM went ahead and opensourced the whole thing.
Now, they COULD opensource SOM and (with Apple's cooperation) OpenDoc. That's some old tech that could use a new life.
/Brian
You're missing the major concept: reality.
A) Software has a cost too. Its called man-hours. If you hire a guy to dig a hole, are you going to tell him that you won't pay him, since digging a hole doesn't cost anything? By your reasoning, the book industry shouldn't exist either.
B) Nobody cares about the ethics or reason of being for the software industry. What people care about is that is brings billions of dollars into the economy, and makes the US the world power in computing. Best of all, software is very high-margin, which makes people rich. Rich people spend lots of money. Read up on the basics of capitalism.
C) In an economy, too much money can never be spent. Software is a great place to boost the total money-movement of the economy. Its mostly well-off people (people who can afford computers and software) paying money to other well-off people (programmers.) In the middle, you have tons of people who benifet, and a bunch of jobs created as a result. Software is a big push on the "New Economy."
D) I'm not talking about the fate of a free software projects, I'm talking about the economy. (BTW: Ever thought that having a hardware company support OSS software decreses the already small margins in the hardware industry?) Nobody has proven that OSS is a safe thing for the economy. You could argue that it should be legal for people to give away their software, but a lot of similar things aren't. You can't give away a lot of things. It's a very real possibility that free software could be dangerous for the economy (I'm not saying that it is, it could possibly be.) In that case, it would be regulated just like everything else.
A deep unwavering belief is a sure sign you're missing something...
IBM also makes a lot of money on consulting and integration. That's why they can afford to be so platform-agnostic -- they sell you a solution based around whatever technologies you want, you pay them to configure and maintain it, everybody wins.
(This isn't the way most slashdotters probably like to do things, but a lot of companies rely on IBM's expertise so they don't have to develop any in-house.)
--
Ben "You have your mind on computers, it seems."
All that happened long after the IBM-Microsoft "divorce", and long after anyone had any realistic hopes of OS/2 getting any significant marketshare.
Like most dicussions of OS/2, yours seems to revolve around the period when Windows "Chicago" was very late, there was lots of frustration among PC users, and IBM was making a last ditch attempt at marketing OS/2 directly to consumers. This was around 1993-96. And, it has to be pretty frustrating, IBM employee or not, when you discover something pretty cool only to find out that it was being phased out soon afterward.
The thing to note is that OS/2 already had one foot in the grave at the point PCCO stopped shipping it on every model. IBM had it on the market since 1987, and it never recieved any significant corporate adoption, either as a workstation or a server OS. It's goal was to drive corporate host applications and IBM-style integration, and it utterly failed there after six plus years of trying. Only after the corps rejected the thing did IBM try to save it's hide by marketing it to power users.
Maybe one thing IBM learned from this is that they need to have 'grassroots' support in marketing something like an OS. Linux has it, Windows had it when they defeated OS/2, and maybe OS/2 had it, right at the bitter end. The "top down" approach of early OS/2 marketing was a huge factor in driving the power users into Microsoft's waiting arms.
--
Business. Numbers. Money. People. Computer World.
Short answer: READ THE ARTICLE. Long answer: If you read the story you will see that IBM is being quite honest about that aspect. What they're saying is that their business goals are compatible with open-sourcing most stuff the community could want. So, while the community could ask for something that they would not release because it's not on their interest (and they'll say NO), almost anything of value for Linux will not have that conflict of interests. Since the rest of the article is about how their strategy is to complement Linux and ride the wave, rather than compete with it, it makes perfect sense. Now, would it be in the best interest of Linux to get that Olympics-related source? I would have to say that no, that's not the highest priority. It would be nifty, but on the whole pointless. Would it be in the best interest to get some elements of that software, that would improve some specific tasks in Linux? Hell, yeah! And they say they would probably open-source it. THAT's the point. Whether they underestimate the community's ambitions, that's another matter, and all the more reason to read the damn article before posting.
SGI previously released their FailSafe application monitoring and restart service. Having the Phoenix stuff underneath it and available for the GFS file system, and using the existing linux-ha bits would pretty much be a complete cluster solution. That would be good.
-dB
"It if was easy to do, we'd find someone cheaper than you to do it."
While at an engineering expo at my university, I spoke with a guy who works with Linux at IBM. He said that IBM has been doing lots of testing and r-and-d with Linux. He said that their experts are predicting that Linux will be the dominant OS in 4 to 5 years. He also said that IBM doesn't believe a lot of the current benchmarks with Linux, especialy those with high performance disk access and with SMP, and that IBM is in a position to help Linux immediately, but doesn't want do alienate the community by making it look bad. The guy sounded sincere and seemed to know what he was talking about. I'm looking forward to see what IBM has to offer Linux in the coming years
IBM should be more concerned about forgiving itself about OS/2. Granted, I'm not excusing MS, but part of the blame for the OS/2 failure lies squarely on IBM's shoulder.
Also, if they will support Linux across their entire hardware line, doesn't that also include the Aptiva series? It'd be hard to even RUN Linux on one of those, there's so much WinHardware in them!
"Ancillary does not mean you get to rule the world." --U.S. Circuit Judge Harry Edwards, speaking to the FCC's lawyer
That would be a good explanation for why IBM aren't doing what they ARE doing.
So why ARE they doing it?
I would imagine that IBM look at Linux because it allows them access to the upgrade market from small-medium systems. At the moment they tend to be locked out of it, because of OS incompatibility. If they implement Linux on their hardware they can capture some of the top end hardware market for Linux, and can continue to sell the older O/S as well; those users are locked in. (IBM market share would therefore go up.)
Web server too slow? Get a Linux Big Iron.
IBM probably, kinda like Apple, see themselves as a hardware company. They have a niche Big Iron product to sell; and Big Iron is Big Bucks. Their strategy aligns with that thinking. Probably will work too. It's very clever really.
And it does Linux no harm whatsoever...
-WolfWithoutAClause
"Gravity is only a theory, not a fact!"No matter what the underlying REASON for IBM doing and offering these things, the end result may be the same, if the Linux community works it right.
If they can make a buck off of Linux by doing these things, who cares? If the mom and pop Linux consulting company can do it, IBM has just the same right. I know that not everybody is going to be happy with the underlying reasoning for their actions, but as long as it helps the community, I think people should stop bitching so much.
A slip of the foot you may soon recover, but a slip of the tongue you may never get over. -Benjamin Franklin
Ok so how come there has been a petition for an mwave modem driver (Built in modem on Thinkpads) or specs for it for the last year or more and IBM COMPLETELY refuses to do anything. The might release some of there source code, they might put linux on some of there "cool" hardware, but I doubt they will give us anything we ask for. Cos they have not in the past! And we have asked repeatedly, There are over 900 people on it! Check for yourself here
James
I think when interpreting IBM's stance on Open Source, one should consider IBM's past actions. Originally their mainframe OSes came with full source when you bought it, which allowed a LOT of support by the local sysadmin guru, in much the same way we modify and recompile Linux kernels. There were user orgs like SHARE and people kicked back improvements and bugfixes to IBM, a lot like Linux folk do with Linus. As a matter of fact, a higher performance process scheduler for VM/370 was rewritten by a grad student, and IBM adopted it into the core OS, if I recall correctly.
Then (in the mid-80s I think) IBM jerked the rug out from underneath folks by going to an "Object Code Only" policy, which meant what it said... you get binaries only. The "Just say NO to OCO" button got really popular at SHARE meetings.
They pissed off a lot of people who switched to DEC and Cray and UNIX, and they vowed fervently never to buy Big Blue Iron anymore. Think of how pissed you'd be if Linux suddenly announced he was only distributing binaries. (yeah, the kernel would fork in a heartbeat, but that's not my point, and these folk didn't have that option)
You gotta remember Open Source is a marketing tool and a weapon against Microsoft, in IBM's eyes. I wouldn't trust them not to then pull the same damn stuff they've done in the past.
Not to say their work isn't great and we shouldn't take advantage of it, but just remember Big (friendly 'ol) Blue is still the same big company it always was.
Sure; but I could do something about it without digging into the innards. And I sure could move/deleta any desktop object I wanted, unlike the wired-in inability to delete, say, the IE icon on a WinXX desktop.
I can't argue esthetics are unimportant, but there's nothing more aggravating than a pretty desktop with *NO* functionality, IMHO.
That's it, "delete from users where user_id180000" Hmmm.... wonder how many users over 18k have mod privleges? (:
You'd wonder seeing as I can't even properly use a '<' in my post.
Of course, the poor support and IBM's unwillingness to break previous programs to fix major issues like the single system input queue didn't help matters much.
If you happen to be on the inside, you have access to the IBM internal forums. The OS/2 Advocasy one was about half bitter bitching and moaning about how badly IBM fucked up with OS/2 and about half Linux advocasy. Many of the OS/2 people I knew inside and outside went to Linux after it became apparent that IBM was never going to do what needed to be done with OS/2.
In a way it's better. IBM doesn't own Linux so they can't screw it up like they did with OS/2. It's not nearly as succeptable to the FUD as OS/2 was, and it's not being maintained by a bunch of people who think PCs are toys that you use as dumb terminals to the Big Iron.
I'm trying to teach myself to set people on fire with my mind... Is it hot in here?
This has been endlessly discussed in the comp.os.os2.* newsgroups. The real question is whether the System Object Model can be ported off the platform - especially since IBM has now essentially dropped all development of the technology. The consensus (granted, of outsiders) is that trying to do so would be a nightmare.
We can all dream, though!
Are you implying that Cisco doesn't try to control their market? I'd have to disagree. They have a reputation for throwing their weight around, though I admit I don't have any examples ready to hand.
--
Ben "You have your mind on computers, it seems."
But IBM looks a lot like Cisco - sure, they've got patents coming out of their wazoo, but they don't use their strength to control the market(so it seems. One can't be too sure).
:p
The Justice department already called Big Blue on that tactic..
Your Working Boy,
Basically, I don't think you realize how intimately entwined parts of the kernel are. Even if you put the "big iron" code in separate files, changes elsewhere would break it every minor release.
The evaluation of an action as 'practical' . . . depends on what it is that one wishes to practice.
I wouldn't be suprised if Microsoft still has legal rights involving the Workplace Shell (OS/2's GUI) code and underlying technologies.
There's always been this debate about "open sourcing OS/2" (kernel, drivers, Workplace Shell, etc.), but a couple of things always come up whenever I've heard it discussed:
1) The code is supposedly very messy.
2) It would be EXTREMELY difficult to figure out what could legally be released to outside people. Who's going to pay to do that?
3) If you can't release it all, what use is it to people? What incentive does IBM have to do it, other than "it would be nice?"
BTW, I still continue to use OS/2... my customers still want me to support it.
Recently there were several articles in /. about Linux not supporting "big iron" features. Probably it is true. But in this case why doesn't IBM support also HURD?
HURD has some fetures, which are perfect for "big iron" and are unpleasant for smaller machine. Why not help to make HURD 100% Linux compatible and use it on big iron, clustering,.....? T
his would be a real BIG thing for IBM and will prove they support open source. For now it is simply "riding the wave" and lot of words.
That was a mean trick Malda. Why did you have to remind us of the good old days. No Trolls, no hot grits, no Natalie Portman -- when /. accomplished something. When it DID something. Now it is just a recycler of news -- which I like, but it used to be amazing. You know what I mean.
*sob* *sob*
-Davidu
# Hack the planet, it's important.
There's an editorial about the topic "IBM and Linux" in the current c't (20/2000).
A monkey is doing the real work for me.
I didn't actually READ the story, but I skimmed it and the quote about "we'd pretty much release anything the Linux community wants" (which was in big letters in the middle of a page) jumped right out at me. "Uh oh", I thought, "when Slashdot runs this story (which they are guaranteed to do, if only to get people to buy the mag and thus see the interview with Taco and Hemos) it's Abuse City".
For instance, how about this, IBM: Release the software you used to run the Olympics.
Now, I'm not accusing IBM of jumping on the bandwagon--I think their Linux support has been great (practically) and genuine (philosophically). But that doesn't warrant crazy statements that imply Gerstner is ESR or RMS's bitch. For good or for ill, IBM's goals do not correspond with mine--therefore not everything they do will make me happy. They'd be more honest (and thus keep their good rep clean) if they just admitted this.
--
Non-meta-modded "Overrated" mods are killing Slashdot
(Hey Ryan! Here's your proof!)
When Linux Mag notes that Linux is more a "development methodology" than an operating system, Wladawksy-Berger neatly sidesteps the issue, merely responding that it's important that developers can easily port their apps between Linuces. (He does add a bit of abstract and highly-qualified praise for community-based development.) Truth is, IBM won't change their internal design lifecycles that easily, and, for the most part, I don't think they need to.
....
Nonetheless. This is exactly what Linux needs: a high-end, organizationally-driven addition that will help make Linux a viable choice in the enterprise arena. For the first time, I'm starting to consider Linux a real competitor for my business. I'm not ready to turn in my E10K yet, but
Did anyone pick up an undercurrent of glee when the dude compared who uses Windows (half the class raised their hands) and who uses Linux (the whole class raised their hand).
MS should never have fck'ed over IBM wrt to OS/2, Big Blue has a long memory, and the board probablt has wet dreams of stomping on MS's grave.
HOW COULD YOU SWITCH FROM OS/2!! OS/2 was my baby. I'm gonna go weep in a corner now. OS2 WILL RISE AGAIN!!
Does anyone know of any technical reasons why you can't just have some kernel option 'big iron' that switches all the right things on when you want them?
dnnrly
The IBM developerworks page announces that IBM will open-source AFS later on, which is far more important than their open-sourcing of JFS, imho. While ARLA exists and is cool et al., the real thing is still a notch better.
IBM is growing in my regard every second. Now if they only sent me some old hardware with the MCA-bus... :^)
Ok, I'm missing something here: What's the big deal? I mean, Linux is cool and all but why bother if the thing already ships with a UNIXy OS already designed to run on that machine? z
This is the windowing system/user interface to OS/2. Compared to Windows, X-Windows flavours, Mac, and maybe even BeOS, it is the best windowing interface I know. I wrote a multi-media tourism application under OS/2 for kiosks, and I had full-motion Mpeg video, sound, effects..and that was in 1992! It's only now that Windows is catching up. WorkPlace Shell, WPS, push for it!
ttyl
Farrell
Former OS/2 Developer, using AVC
CAN-CON 2019 - Ottawa's only book oriented Science Fiction Convention! October 18-20, Sheraton Hotel, Ottawa, Canada h
Makes you wonder if Big Blue is still pissed at MS for all those dirty tricks years ago, now doesn't it... :)
NecroPuppy
---
Godot called. He said he'd be late.
I like you, Stuart. You're not like everyone else, here, at Slashdot.
Why not have them opensource all the parts of OS/2 that they legally can do so and the entire API so that Linux can be made to run OS/2 programs the way it is trying to do for Windows with Wine. Of course code from the GUI engine could move towards merging w/ X and/or forming an alternative GUI for Linux.. mmmm OS/2 Tux.. finally a new version of OS/2. :)
At what price learning? At what cost wisdom? The price is a man's peace of mind, and the cost is his life.
This interview underscores the fact that "embracing open source" for most traditional UN*X outfits means that Linux is labeled a low end solution while offering an alternates for High end applications.
It seems that Linux is adding to the profit for these companies because they can sell more low end machines with an OS that more and more people are familliar with. This saves them development time and programers because most of the OS is open source. They also keep their high end UN*X for the Big Iron or special computing problems(like beating a chess champion or folding a protein).
Interesting business plans with the OS community doing a lot of leg work for the higher volume lower priced units. Is this the wedge that many OS companies are looking to drive into Microsoft?
"Science is about ego as much as it is about discovery and truth " - I said it, so sue me.
>There is also some talk of actually having a fork in the code so that there can be a big iron Linux and regular Linux, by the way.
Na, they won't fork the kernel. There are already kernels I've seen (linux-vr) that just have added classes, PDA in that case. All one would need to do is merge a BIG_IRON class in with the standard kernel. It all gets handled in the config, and then with #ifdef's. No permanent fork necessary.
The poster you replied to raised a good point though. ever though the "Big Iron" support would bog down 486 installs, it should (hopefully) be possible to have configure/compile time options that would completely remove the big iron code from the compiled kernel and you would basicaly have the 486 version running as fast as possible.
;)
There really shouldn't be anything wrong with that (95% of users never use anything but their distro's stock kernel anyway, which would have the big iron support removed before they touched it.)
Code forks are dangerous things when they are done for reasons like this.
From what I understand from the kernel Mailing list and others I have spoken with, the reason big iron support is not getting into the mainstream parts of the kernel is because IBM has not made it a configurable option - it's been proposed as schedule and memory changes in the main parts of the kernel which I agree would be bad.
I don't think it will be long before IBM changes the policy and it' sall good
The GPL & LGPL make it very difficult for IBM developers to work with both LINUX & AIX. I think it's great that they're expanding into the LINUX arena, but as was demonstrated by SUN recently, anytime a company tries to break into the community with some open source program, all the open source geeks pounce on them, claiming their entire OS/Life's Blood is now free. The number of lawyers needed by any company that wants to make money off licenses but also wants to get involved in Open Source just sickens me.
--
You, you and you: panic. The rest of you, come with me.
I'm really easy to get along with once you people learn to worship me.
This has already discussed in a previous story today, and nobody came up with a satisfactory answer as to why there can't be 2 memory subtrees, one for big iron and one for desktops with the same API. I know that it's duplication of effort, but considering that the kernel is extremely well managed at the moment it wouldn't be as hard as it would be in a corporation where every manager has their own political agenda. It's obvious that there is a huge difference between the IA32 and S/390 architecture and different issues apply. It does make me smile though, to think of my favourite OS running every computer on earth at some point, especially when Bill Gates, at the launch of Windows 2000, stated that his product was now ready to take on mainframe-type jobs. That was probably the funniest IT-related statement I've heard this year.
This is a reasonable position, but I'm not sure Linus would go for it. It may dilute his focus so that he can't maintain either one well.
The evaluation of an action as 'practical' . . . depends on what it is that one wishes to practice.
These are interesting times, my friends. Another clash of the Titans. I just read today about M$'s DataCenter, a big-iron release of its Windows 2000 Server. will run on machines with 32 processors and loads of RAM. It's supposed to be Microsoft's best challenge to date to *nix systems. IBM should get linux compatibility and get it quick!
I need a TiVo for my car. Pause live traffic now.
"To be honest, we'd open source just about anything the community wanted."
Then open source the GUI engine for OS/2. This would give Linux an alternative to X, and one that is less resource hungry.
My question is simple - does this cheif strategist primate use Armed Linux or his native Monkey Linux?
click here to incinerate homeless people
There is also some talk of actually having a fork in the code so that there can be a big iron Linux and regular Linux, by the way.
...Open Source isn't the only answer -- but it's almost always a better value than the alternatives...
This is a good point: IBM hates Linux as much as it hates the Internet because it can't profit from it. However, big blue can do something to Linux that it can't do to the internet: package it and slap on it's hyper-rasterized logo. Looks like we can expect this "pig-iron" version of Linux very soon!
"Ancillary does not mean you get to rule the world." --U.S. Circuit Judge Harry Edwards, speaking to the FCC's lawyer
Interesting. Very much as I thought. I guess the frustration is all-the-higher because at the time (1990) MS was the little guy and IBM was still Big Bad Blue.
So IBM really couldn't gain any support even though MS was in the wrong.
Well -- what goes around, comes around.
Okay, so I don't buy millions of dollars worth of hardware every week. I may not arrange licensing agreements with IBM. But IBM looks a lot like Cisco - sure, they've got patents coming out of their wazoo, but they don't use their strength to control the market(so it seems. One can't be too sure). The fact that they're willing to GPL stuff sort of supports that idea.
Dave
'Round the firewall,
Out the modem,
Through the router,
Down the wire,
Barclay family motto:
Aut agere aut mori.
(Either action or death.)
"IBM sees linux as a disrupting technology of the same class as the internet: an OS that can run on many platforms and that nobody owns: something that can fundementally change the landscape of computing. "
This whole think stinks of Katz...
;)