Slashdot Mirror
IBM to unveil more Linux plans
Over at Financial times, there is a story about IBM which will unveil a Linux plan to invest 200 million dollars, helping companies to write Linux applications. Definitely worth a read. Thoughts anyone?
← Back to Stories (view on slashdot.org)
Its non-commercial nature, however, means that no money is invested directly in the development of Linux, in contrast to competing products such as Windows, and the Unix operating system developed and sold by IBM, Compaq, SCO and others.
I'm glad that IBM is throwing this much money into developing apps and support for Linux, but I feel that they could also do something to help in the development of the kernel as well. It almost seems that, with this statement, that they are saying "Someone else can worry about that". Like they are taking advantage of the open source community.
Ok, I'm not totally against this, and this is a statement from the reporter and not from IBM themselves. But I think this is a point that most commercial companies are missing. It is actually to their advantage to offer some expertise to the free stuff. For one thing, it makes you look good in the eyes of the community (SGI sees this). And another thing (which RMS probably won't agree with), is that, by doing so, you can have more influence in the decisions that are made.
Steven Rostedt
Steven Rostedt
-- Nevermind
Nop.
/. for 3 years now - and an author since last year :)
I've been reading
Hetz (Heunique)
Upgrade to Netscape 4.74 (which appeared yesterday on Netscape FTP site), and enable javascript - this should do the trick (at least at worked here on my 2 Linux machines [Redhat 6.2])
Hetz (Heunique)
`competition protecting' legislation: the notorious `anti-dumping'
legislation of the 80s was directed at foreign manufacturers who were
alleged to be selling their product at less than cost. One of its
Keffects were that companies used it to gain injunctions stopping
imports, saying `well, *we* can't make it that cheaply'.
Still I think the story you give is how it *should* work: if you
aren't a monopoly, you should be free to set your prices however you
like, for whatever reason you like. If you are a monopoly
lightly.
I'm just waiting for IBM to release a baseball Game for Linux/X. Then I will be convinced Sisko is running the place.
--- Justin Dearing http://www.justaprogrammer.net/ We're just programmers.
Nearly 50 fresh-faced engineers and entrepreneurs in San Jose and Cambridge work alongside IBM's sharpest minds on newfangled products and services, such as Linux systems management and pervasive computing devices. IBM employees manage the youthful groups.
Okay, so we probably wouldn't call Linux "newfangled", but it's food for thought for all you college seniors who want to work on open-source, and get paid.
I'd rather have someone respond than be modded up.
IBM's strategy is very smart--help APP developers start using Linux and there will be more apps--which means more users, which means more of everything for everybody.
But throwing money at the kernel people gets you nothing. The kernel people aren't driven by money. You might conceivably find someone who was unable to implement a feature due to lack of money, but all your money has bought is the feature--not apps that exploit the feature.
In short, IBM is throwing money at problems that benefit from having money thrown at them--but no farther.
--
Linux MAPI Server!
http://www.openone.com/software/MailOne/
(Exchange Migration HOWTO coming soon)
IMHO, throwing big bucks at Linux is an attempt to turn Linux into a system that IBM can use to blow away Sun and DEC at the low-end, and boost their sales & support.
NOBODY spends money, unless they feel they're getting -some- return, of whatever kind that means something to them. In IBM's case, long-term survival versus "budget" systems (such as Wintel) and short-term improvement in relations with geeks and companies make for some plausable return that IBM might well want.
IMHO, also expect to see IBM and SGI work jointly on Linux. So far, both have been doing a lot of work in EXACTLY the same areas (journalled fs, Apache improvements, etc), and so it would make a lot of sense if they were to combine resources, rather than duplicate work.
It's a small world and it smells funny; I'd buy another if it wasn't for the money; Take back what I paid (SoM)
Microsoft gave us the answer: look at their microsoft.net/whitepaper, read how they plan to change the world, and then, let's sit together and thinker how we, the people, can make something better, something worth living. We need free, standardized solutions for the following areas:
- A Privacy/Identity Framework
- Building Blocks for Net-Applications
- Governmental Tools
That's a lot, I know. And all the Biggies (Microsoft, AOL, Nokia, Media, TelCo's, etc.) will be fighting for defining the standards of the tomorrows net-world. We have to act now, in order to define our standards - you do know why, don't you?A Client/Server solution for identifing people in a secure manner, where *we* control our privacy. This is a must for serious e-commerce, e-governemt.
Tomorrows Net will be in the center of tomorrows society. Connected by mobiles, and a myriad of other devices, we will communicate, deal and live together digitally to a great extend. And we do not want to let a single corporate (or governmental) entity control how things will work.
The more we "live" on net, the greater will be the need for some control - that's what the government was for the old economy. In the new economy we want to define how this control works, who controls whom, what's right and what's not. This all depends on the infrastructure.
I agree, but instead of deciding on the libraries themselves, how about standardizing the interfaces instead? This will allow anyone to create a library in whatever language they choose. I cannot see an easy way to create a common set of static or dynamic linkages to satisfy all languages, but you could standardize socket interfaces. For example, a math package daemon could sit out on port 9000 (e.g.) and take commands to do things like complex math, matrix math, etc. Then a plotting package sits on 9001 waiting for a 2-D data stream, 3-D data stream, etc. Don't like your plotting package? Download and install a different one. This is a lot of what enterprise java beans (EJB) are trying to do.
The cool thing is that, if you want, you could publish these services to allow others to use them on your box. App servers anyone?
I can see how the network overhead might be a problem, but if you only need localhost support perhaps a virtual network driver (similar to what VMWARE does) could virtualize the requests so they are all software and, therefore, much faster than an actual network socket call.
-tim
Microsoft should never have gotten IBM angry over OS/2... I am sure that for IBM, revenge is sweet.
In the late 80's and early 90's we ran networks of AS/400's that had better uptime than our commercial MVS environment. Exactly 0 unplanned IPL's in 8.5 years. Only downtime was OS upgrades and tape drive replacements. Rock solid the most reliable stuff I've ever seen ever better than 4381's. But remember some of that came with operational discipline, planning and rigor we borrowed from the MVS world and its 30 years of refinement.
Ibm want to use linux so they can sell more Net-Appliances. That is why they are doing this. Which is not a bad reason. They want to sell more iron they could not care less about the OS that hardware runs on they just want to sell it.
While I'm sure revenge is on the minds of those in charge (and mine actually) but IBM wouldn't waste millions for such as a lowly purpose. This type of revenge is what makes them happiest, a step into the future.
AIX (and the others) has been floundering for years. IBM has by this time put more money than most of us can imagine into developing UNIX/AIX software/hardware. We (should) all know what MS did to them.
Now comes an opprotunity to:
1. Regain a use for their most of software and hardware.
2. Seek revenge on Microsoft
3. Gain popularity with a community that used to hate it
4. Be one of the firsts in the market
5. Have all future developing & testing done free/cheap
Personally I wonder why more *IX companies aren't doing the same things.
Devil Ducky
Devil Ducky
MY peers would get out of jury duty.
Only problem I see with that is IBM will surely want to buy a well-known distro (if they didn't make their own). In the US the obvious choice would be Red Hat, some other options could be SUSE, Debian, Mandrake, and I don't think Slackware is a viable option for this discussion.
:-)
But for discussion sake they choose to buy Red Hat. They take last week's earning from Thinkpads and buy up all of the stock. No more Red Hat bumper stickers, no more pictures of penguins wearing red fedoras (always looked to me like a penguin that just got laid by Carmen San Diego). But now they also control GNOME.
I believe that the GNOME/KDE debate is one of the best things happening to the linux world. It spawns better graphics, stabler desktops, more built-in features, more customizationabilty, etc. Who knows what would happen in IBM's hands?
Sorry, I forgot where I was going with this, just make up whatever ending you want. All I ask is that in your ending I sound smart.
Devil Ducky
Devil Ducky
MY peers would get out of jury duty.
There's no serious doubt that anti-dumping laws were widely abused by
US companies. Frequently the whole aim of pleas was to obtain
`short-term' injunctions, and then attempted to drag out for as long as
possible a case they couldn't win. Whatever your take on GATT and the
WTO, they have at least cleaned up international trade law a great
deal.
I've had too damned many denial-of-service attacks disguised as JavaScript pages to allow it to be enabled on any of my machines.
And FS does not have a proper "webmaster" feedback address--webmaster@fs.com bounced. MAY THEY ROT LIKE THE BASTARDS THEY ARE!!
"My opinions are my own, and I've got *lots* of them!"
Read the article again. This is just the beginning..
Traditionally, Linux is much more acceptable in Europe rather in US or Canada (don't take my word on it - Ask IBM Europe, or SAP)
Hetz (Heunique)
--
-- Slashdot sucks.
Simple - IBM sells Big Iron. If Linux becomes everyone's OS, IBM's $200 million will be an investment in DISPOSING OF THEIR SOFTWARE BUDGET. And it will also take a big chunk out of their support budget. IBM makes money off hardware, period. AIX/390/other OS's are just something to put on the hardware.
It's rare that you're presented with a knob whose only two positions are Make History and Flee Your Glorious Destiny.
But hopefully people won't consider IBM to be an Uncle Tom because of this.
If you haven't seen the ad, btw, here's a transcript:
I just love the way he pronounces "software" - it's sexy. One thing though: how come he pronounces Linux as
LILO boot: linux init=/usr/bin/emacs
The press release from IBM is here: http://www.ibm.com/news/2000/07/21.phtml.
More interesting is the page at http://www.ibm.com/developer/ linux/eu_en/program.html which has a little more detail.
---
"Where do you come from?"
Hi!
--
-- Slashdot sucks.
Cross-platform is hard to do. Java is a 'sporting try' at it, but it's still not quite there. With c-p, you necessarily take a penalty hit with a VM, or platform-specific abstraction hardware. It's the nature of the beast, as long as there are different platforms, there will necessarily be platform specificity. I'm all for some sort of meta-assembly language that everyone would agree to implement; but how do you propose to do that?
:)
Portability has it's own issues. Different platforms exists beacuse there are some problems that are better solved one way, and others that are better solved another way. Some chips are good for real math, others for integers, others for parallel matrix work, others for graphics processing and still others exists solely to support LISP. True application portability would require a standard set of libraries (native of course) for the 'portable' language to hook into. (Umm, Java again, some C/C++, FORTRAN have already tried this). And you still have to recomplie (though not necessarily edit much code) to run on the new system - and there's keeping the libs up to date to worry about.
We just push the problem to another layer.
Asking for the problem to go away isn't going to do it. There will always be cross-platform issues - for as long as there are different platforms; and for just as long, there will be portability issues.
Many people have been trying to solve the problem of multiple platforms, and all of them have developed pretty good solutions. Sun, Apple, Intel, Microsoft, Linus, IBM...
-- What you do today will cost you a day of your life.
IBM giving money to some "unknown" to develop software for the system they were building...
Whatever happened to them?
----------
Stupid sexy Flanders.
I wish the AS/400 wasn't so hard to learn.
There are a lot of really cool things about it, but I have never seen such a vertical learning curve in my life. You have to understand the whole gestalt behind the system before you can write 'hello, world'.
So Linux on the hardware might indeed be a dream solution.
D
----
And IBM's journalling filesystem is what, chopped liver? Sure, I'd like to see them make more contributions to the kernel, but they've already headed in that direction and purportedly plan to do more. IBM's plans to port Linux to the AS/400 and S/390 architectures will almost certainly result in kernel code!
While it would be stretching the truth to suggest that I fully trust IBM, a number of high-level IBM execs have pretty well staked their careers on the success of Linux and free software in general and IBM has made some significant contributions to the body of open source software. IBM's open source Java software goes a heck of a lot further than any gesture Sun has made.
Besides, while I'm sure Slashdotters all have a wish-list for the kernel, the great unwashed masses currently sucking at the great tit of Redmond don't -- they'll plainly accept any old operating system, but they want apps. Heck, I want more apps.
Proud member of the Weirdo-American community.
The immediate question I ask, when I see big numbers like "$20 Millin" thrown around, is *what are they after?*
IBM isn't doing this out of the goodness of their hearts. They expect payback.
They expect to get their twenty million back PLUS another twenty to sixty million minimum.
So where's the money going to come from?
--
--
Don't like it? Respond with words, not karma.
They need to bring back Denis Leary for their IBM/Linux commercials. He'd make the perfect outspoken Linux zealot
"Microsoft!?!? Fuckin' Bill Gates!@#!"
So you should be quite pleased with this announcement!? These centres give developers the opportunity to try out their applications on various Linux distributions and AIX, and on various hardware platforms. You can even test your systems remotely over the internet without visiting the centre.
To my mind this is a great step forward. How many small Linux development organisations have the resources to really test on different hardware or even software distributions? With this announcement they have the ability to test and produce the "cross platform software" that you desire.
Sure, I'd like to see even more unixes and more hardware (Sun, HP, Cobalt, ...) but that is probably a little much to ask from IBM. Let us instead hold up IBM's initiative to the other vendors, and try to persuade them to launch something similar.
As for "Linux software": I guess the commercial realities of today is such that you have to include the word "Linux" or e-something-new in a press release to (a) get any attention on /. (and the rest of the media) and (b) get your share price going.
This really is good news, if the access is releatively open and free.
---
"Where do you come from?"
Hi!
fwiw, I'm running NS4.71/Linux/noJava/noJavascript.
"My opinions are my own, and I've got *lots* of them!"
Big companies like IBM and Sun "give" stuff away because they expect it will help them sell more stuff in the run.
* We GIVE you the software and you BUY the hardware
* We GIVE you the hardware and you BUY the software
* We GIVE you some hardware and software and you BUY some support
Who ever gives away the least and gets the most wins. End of story.
It has NOTHING to do with Linux / Open Source philosophy, so don't whine when the Financial Times does not appear to "get it right".
IMHO Microsoft has been wooing developers for years. In addition to the stuff they sell, they provide free SDK's for all sorts of technologies:
Winsock, TAPI, DirectX, Agent, Win32, OLEDB, MAPI, Plug & Play, etc.
These APIs may be closed, but they drove a few million developers into the Windows camp.
A few development/porting centers is NOT going to make up for this. Sun used to do the same thing 10 years ago. It didn't work and they are still trying to figure out what they can give away for free to convince people to give them money.
I see lots of argument still over windowing managers, distros and packages -- how about common things like DDKs to encourage hardware vendors to support Linux.
Is this sig nificant?
Not that I know of, but I really, really hope it is. That would do wonders for the adoption of Linux by businesses. IBM is a big name, and technologically challegened management types tend to do what the big names tell them...
"The invisible and the non-existent look very much alike." -- Delos B. McKown
I'd still like to see an IBM distro if for nothing else but to legitimize it a little more to the PHB types or software developers who are hesitant to port over to it. A big well-known name like IBM's on a distro would help things a lot.
It might be expensive, but what if they just bought out RedHat or Mandrake and renamed it "IBM Linux" ? (or IBM RedHat, or whatever)
It could be a smart move for them, but it also makes me wonder if they are doing this because they are still a little bitter from the whole Windows-vs.-OS/2 thing. :)
Information wants to be anthropomorphized.
It's nice to see some of that money going back to the motherland, where it will be appreciated. I'll pretend that IBM is thanking Europe for developing Linux in the first place--except that I don't see Finland mentioned. :|
However, it's good to see that Intel is in on this one, too. Anything they can do to annoy Microsoft always entertains me.
Now, I don't expect a great degree of technical accuracy from the Financial Times, but I always snicker at that "running webservers" stuff. I guess that's all people care about. Forget that mundane crap like DNS, Mail, News, Timeservers, Database Servers, NFS, FTP Mirrors... All we know about is the web. Web pages, yeah, that's the ticket.
I'm not even going to mention compilers, image processing, clustering... I mean, really, who cares if it's not on the Internet? And if it isn't on the web, well, where can we find it? Isn't the Internet the web? Isn't that AOL? Ah well...
---
pb Reply or e-mail; don't vaguely moderate.
pb Reply or e-mail; don't vaguely moderate.
I don't mind software that's tied to a particular platform. While portability across platforms is a good thing, it isn't essential. I don't have any problem using software that's only available on one platform.
The problems, IMHO, are caused not by single-platform applications, but by single-application data formats. Software is just a tool for manipulating data. It doesn't especially matter which tool you use on which platform to manipulate a particular set of data, so long as you can transfer that same data to another platform and not be SOL. MS Office isn't bad because its only available for Windows, but because it works with data formats that aren't readily exportable to other platforms or applications.
Having said that, I do recognize the value of an application that runs across multiple hardware and software platforms. But, I don't see cross-platform software as being nearly as important as cross-platform data.
There's no such thing as Scotchtoberfest!
If they gave a few S/390 boxes away to developers to use, we'd see quite the proliferation of Linux for S/390 software. That can only help Linux and IBM. Or, maybe they could do a test-drive program similar to what Compaq offers, basically you sign up to get an account on one of their machines running various flavors of Linux.
For those of you who might otherwise enjoy dissing this established hardware, make sure you know the facts:
- There are millions of 'em in the world
- In many environments management would appreciate the ability to migrate to something more unix-like
- These things have awesome I/O bandwidth
- 65,536 IRQ's!
I've personally witnessed the first three points. Especially the third one; Linux on an LPAR boots in under a second. I also saw it dump 17,000 files out of a 160mb tarfile in 3 seconds (!). Now that's just damned fast.A Linux on 390 guy,
DragonWyatt
Don't sweat the petty things. But do pet the sweaty things.
...then they should offer everyone the chance to buy their IBM computers with Linux preinstalled. I was looking at their site the other day, and they have some cool small-form-factor desktops. I was thinking about getting one for my 12x12 dorm room, but I'm gonna have to pass because I'd have to buy a Windows license (which I neither want nor need) with the computer. You may think that I'm being petty, but I am on a tight, tight budget and the price of a Windows license is a big deal.
I think it sucks that these corporations -- while paying lip service to Linux, Free Software/Open Source/whatever, and consumer choice -- still force you to buy crap like Windows, Office, and God knows what else for each system you buy.
========
Stephen C. VanDahm
IBM has been pleasanly surprising us ever since they made Benjamin Sisko their Captain. Coincidence? I think the facts speak for themselves.
IBM Hardware:
It might not be the fastest, quietest, smallest, best looking, most cutting edge -
But by God is it reliable.
Once worked for a company with 43 IBM AS/400 machines, one at each of their sites. We had a disk crash about once a month.
Bad? They were all over 10 years old, and had NEVER been rebooted, or turned off. (Oh - and we never lost any data, the diags built into the hardware gave you just enough time to pipe the data off the disk before it went bang).
I would like an industrial IBM machine with Linux please - have it oiled and sent to my room immediately.
When I were your age, all round here were fields...
Linux, a rival to Microsoft's Windows that is free to use.
Linux, the brainchild of Linus Torvalds, a Finnish computer scientist, is non-commercial. It is free to use...
No mention of the GPL, or the philosophy behind it. No mention of RedHat, or S.u.S.E., or Caldera.. Yes, the kernel is non-commercial.. But the article is misleading. When will journalists get it right?
wishus
---
Compared to what it cost them to have Avery Brooks talk about the lack of flying cars on prime-time TV several times a week, $200 million is a drop in the bucket.
Information wants to be anthropomorphized.
As a person who used CICS to connect S/390 to NT using CICS, I PRAY that at least the CICS gateway would be ported to Linux.
:(
At my previous job, they were simply "Anti Liunx" and I had the single chance to install Linux and show it at work - I found that CICS is not available for Linux
Hetz (Heunique)
I spend years telling people to write cross platform code and now we get companies like IBM promoting Linux software.
I hate Linux software, I hate Windows software I'm fed up with this 'one OS to rule the world' crap.
I want cross platform software. Anything that says it works on Linux but not other unixes I don't touch with a barge pole. I get fed up with crap like "KDE, the popular Linux desktop". KDE runs on multiple platforms dammit, as does Gnome, Apache and all the other poster children of the so-called Linux revolution.
I have no interest in Linux software, only good, cross-platform free software.
Grrrrrrrr.......
-----