I worked for IBM for 16 1/2 years. I don't recall having my time tracked, writing weekly status reports, or screwed out of comp time. As long as I delivered, they did not bother me. Of course, I had to write reports and presentations, but when they were needed, not weekly for the sake of weekly.
This true at least for the Gerstner era for me.
The older IBM was more like you describe.
depends on what "enterprise" means here.... i'd be skeptical of Postgres. But, now we have Inprise.... and yes we would have to scope the problem better. But support issues are easier to deal with (with skills and using popular OSS where there are multiple choices of support) than with even IBM or Sun closed SW.
I'm kinda wondering about BEA's long term business model given they are the high volume leader, but not revenue leader, e.g. by my estimates, WebSphere is more than 2x BEA, the company, in revenue.
But much worse, will Enhydra, etc... do to them what Linux has done to SCO? Or Apache to IIS or Netscape?
Pretty lame posting this as anonymous coward. I've seen technical and functional studies that illustrate that AIX has the richest function set of any UNIX/Linux/BSD/NT.
However, if it is HA clustering related as one can glean from the note, then there are probably multiple opportunities as I mentioned, unless it is a solution to a trivial problem that requires none of that... in which case, use the MySQL business model.
It depends on what this SW is and does though. Sounds like server-side SW, probably in an HA cluster environment?
1. Subscriptions to enhancements 2. Technical and how-to support 3. Consulting 4. Add-on or related products...e.g. hardware, other SW, books, etc... 5. Build a total solution around the OSS 6. Advertising revenue on the web site (assuming volume traffic of course) 7. Use the old MySQL model of keeping the newest release closed or gated and fully open sourcing the previous release
Are we building a pyramid? What is to protect it from 10,000 years of natural and perhaps man made disasters? If it is not in an "obvious" structure or monument, how will people or whoever find it?
Very few of the buildings or storage facilities built in the last few hundred years are likely to have lifetimes greater than 2000 years or so.
Also, I would not assume a linear or exponential progression of technology over the next 10K years either. If one looks at history, for every two steps forward, humans have taken anywhere from 1 to 3 steps back depending on time period. e.g., the Egyptians (3000-1000 B.C) and the Romans (500 B.C. - 500 A.D.) had better technology than Dark Ages Europe (500-800 A.D.). Only in the past 500 years have we advanced steadily with respect to technology.
He's potentially right from a traditional mom and pop small business and branch office market point of view.
OTOH, SCO has virtually zero presence in the Internet market....so this will not get them there.
BTW, I'd guess there are 2 million SCO servers installed and perhaps 2 million Linux servers installed. But Linux is growing *much* faster. And Linux is eating into their install base. This is a defensive move and quite necessary for them to stay alive.
Now did their market cap go to >$3 billion today to be like Red Hat's? That will tell you what the Wall St. boys and girls think.
The air pressure on Mars is about 1/200th that of earth. So even if we convert the entire atmosphere to O2, it still is very unbreathable.
Then there is the little problem with UV and cosmic rays from the sun which are not screened out as they are on earth e.g., no O3...etc...
Finally, we do not know if there is microscopic life on Mars that is buried or dormant. If there is, we don't know what would happen if we "woke" them up or made the surface more hospitable. Especially if we accidentally brought them back to earth.
They can't: too much 3rd party code including MS code that is legally incompatible with the GPL or any OSS license. Some pieces of OS/2 could be opened. In fact, I believe IBM is doing that...see their site. Too bad though, OS/2 is still well ahead of Linux for desktop functionality and useability (e.g. WPS vs KDE/GNOME).
will it be commercially sucessful. remember NT on RISC? it is an interesting reinvigoration strategy for S/390.... just got to fire up that S/390 in my basement and get to work on it. Or maybe IBM will give me one.;-)
Linux is not supported and will not run on an RS/6000 S80. I believe Linux us only supported on low-end RS/6000s up to 4-way. The Linux kernel needs lots of work to support a 24-way SMP system.
Good article. But a bit misleading. IBM had a Linux strategy that was announced in March, 1999. This recent announcement builds on that work of 1998/early 1999.
IBM's Linux strategy development got underway in 1998 not Oct 1999.
This is server SW license sales. Which includes preloads, CDs and upgrades to existing installed Linux server SW licenses. Does not include copies or downloads or throw-aways.
Given Linux's business model, one should not expect much revenue from the OS...ever. (ref: Bob Young's goal about collapsing the OS market to $500m from several billions).
What is more interesting is how much new HW (units and revenue) that goes with this. And of course, these unit numbers are great news for SW middleware providers and services companies as it represents a great expansion of opportunity (and no MS and BackOffice to compete with!).
I agree with some of the OSS points in the replies here so will not embelish on them. But I do want to make another point.
I'm walking the floor of LinuxWorld and seeing all of the excitement, etc....It's fun, lots of cool stuff, etc... Reminds me of the old USENET and Uniforum conferences in the mid to late 1980s when UNIX was the next big thing.
Yes, some of those companies died or if luckier were bought at a nice price (Apollo comes to mind). Some though went on to become very successful (e.g. Sun, Oracle) *despite* the IBMs and the DECs of the world getting into the UNIX world.
I'm not sure what the lesson OSS leaders are going to learn as implied by this article is, but it certainly appears that they are in a much more upbeat mood and excited and having more fun with cooler SW than say the Windows 2000 development team?!? Even if Win2000 is not a "train wreck" as ESR says, many in the development team must be so burnt out by now and depressed from all the negative press that *they* will feel like they were *in* a train wreck.
The OSS leaders do not have to fear this fate for reasons stated in other replies.
Army of Northern Virginia
Sun has a significant business problem with Linux
on
Free Solaris 8
·
· Score: 1
Sun does not sell Intel HW, which is what >90% of Linux is based on. Sun does not have a sigifnificant revenue/profit producing SW business nor services business. So, Sun has really no way to make significant money with Linux except perhaps to sell large scale Sun servers as a database server behind many smaller Linux web servers and things like that.
IBM has an Intel HW business and SW and Services.
So, yes, Sun will be on the defensive and will probably have to incrementally change its strategy over time.
Can you say "Let's build Intel servers", Scott?
Army No Va
Re:AIX vs Solaris vs Linux?
on
Free Solaris 8
·
· Score: 1
AIX's origin was System V.1 and Interactive starting in 1983. It first shipped on RT PC in 1986, PS/2 and S/370 in 1988, then RS/6000 in 1990. BSD additions came in the late 80s. AIX was a key foundation for OSF/1, but IBM chose to stay with AIX primarily due to customers not wanting to absorb switching costs.
Solaris was originally BSD based starting in 1981-2. Sometime later, Sun incoporated System V APIs and renamed SunOS to Solaris. Moved to Sparc from Motorola in 1987.
AIX has consistently been a better "enterprise" OS than Solaris as rated by several different consultants. However, Sun has "celebrities", stronger marketing, better visibility and its Internet roots.
Linux will take away much of the low end UNIX server market from both of them due to its business and development models and people with Linux skills coming from school these days.
Several years ago, I was a 2nd line manager in a major UNIX OS organization. Our teams were dissatisfied with the process of working across groups and productivity was low. Along came an ISO9000 "initiative" from above. Rather than grumble about it, we got a team of the top technical leaders as well as some entry people of the organization and let them define how to go about it. The end result: happier and more productive employees, *reduced* overtime and frustration (read....build breaks, breaking more things than fixing, etc...), less stress and gaming for "visibility" or "position", etc...
It's not easy, but it's been done and with benefit. Just have to make sure it does not turn into a beauracratic process, injecting more inefficiency than productivity, though they'll be bumps along the way.
Slashdot Mirror
I worked for IBM for 16 1/2 years. I don't recall having my time tracked, writing weekly status reports, or screwed out of comp time. As long as I delivered, they did not bother me. Of course, I had to write reports and presentations, but when they were needed, not weekly for the sake of weekly.
This true at least for the Gerstner era for me.
The older IBM was more like you describe.
Yep...!
depends on what "enterprise" means here.... i'd be skeptical of Postgres. But, now we have Inprise.... and yes we would have to scope the problem better. But support issues are easier to deal with (with skills and using popular OSS where there are multiple choices of support) than with even IBM or Sun closed SW.
I'm kinda wondering about BEA's long term business model given they are the high volume leader, but not revenue leader, e.g. by my estimates, WebSphere is more than 2x BEA, the company, in revenue.
But much worse, will Enhydra, etc... do to them what Linux has done to SCO? Or Apache to IIS or Netscape?
Pretty lame posting this as anonymous coward. I've seen technical and functional studies that illustrate that AIX has the richest function set of any UNIX/Linux/BSD/NT.
Which version are you using? What are the problems?
Have you considered open source Enhydra? Then you can have multiple sources of support.
What about Linux, PostgreSQL, and Enhydra combo?
Given we don't know what the SW is or does...
However, if it is HA clustering related as one can glean from the note, then there are probably multiple opportunities as I mentioned, unless it is a solution to a trivial problem that requires none of that... in which case, use the MySQL business model.
It depends on what this SW is and does though. Sounds like server-side SW, probably in an HA cluster environment?
1. Subscriptions to enhancements
2. Technical and how-to support
3. Consulting
4. Add-on or related products...e.g. hardware, other SW, books, etc...
5. Build a total solution around the OSS
6. Advertising revenue on the web site (assuming volume traffic of course)
7. Use the old MySQL model of keeping the newest release closed or gated and fully open sourcing the previous release
Good luck!
Army No. Va.
err....10,000 yrs from now is Y12K
Are we building a pyramid? What is to protect it from 10,000 years of natural and perhaps man made disasters? If it is not in an "obvious" structure or monument, how will people or whoever find it?
Very few of the buildings or storage facilities built in the last few hundred years are likely to have lifetimes greater than 2000 years or so.
Also, I would not assume a linear or exponential progression of technology over the next 10K years either. If one looks at history, for every two steps forward, humans have taken anywhere from 1 to 3 steps back depending on time period. e.g., the Egyptians (3000-1000 B.C) and the Romans (500 B.C. - 500 A.D.) had better technology than Dark Ages Europe (500-800 A.D.). Only in the past 500 years have we advanced steadily with respect to technology.
Hmmmm.... as of 12:50pm EDT...
SCOC - 192 million market cap - down 1%
RHAT - 2.8 billion " " - down 7%
CALD - 516 million " " - down 7%
well.....Wall St does not think SCO will take over much... but it is not favorable for Linux companies right now anyway.....
He's potentially right from a traditional mom and pop small business and branch office market point of view.
OTOH, SCO has virtually zero presence in the Internet market....so this will not get them there.
BTW, I'd guess there are 2 million SCO servers installed and perhaps 2 million Linux servers installed. But Linux is growing *much* faster.
And Linux is eating into their install base. This is a defensive move and quite necessary for them to stay alive.
Now did their market cap go to >$3 billion today to be like Red Hat's? That will tell you what the Wall St. boys and girls think.
Army No. Va.
the airlines could make money on those expensive phones in the plane.
The air pressure on Mars is about 1/200th that of earth. So even if we convert the entire atmosphere to O2, it still is very unbreathable.
Then there is the little problem with UV and cosmic rays from the sun which are not screened out as they are on earth e.g., no O3...etc...
Finally, we do not know if there is microscopic life on Mars that is buried or dormant. If there is, we don't know what would happen if we "woke" them up or made the surface more hospitable. Especially if we accidentally brought them back to earth.
They can't: too much 3rd party code including MS code that is legally incompatible with the GPL or any OSS license. Some pieces of OS/2 could be opened. In fact, I believe IBM is doing that...see their site. Too bad though, OS/2 is still well ahead of Linux for desktop functionality and useability (e.g. WPS vs KDE/GNOME).
I've seen S/390s about the same size as 2 to 3 full size PC tower cases....quite small.
will it be commercially sucessful. remember NT on RISC? it is an interesting reinvigoration strategy for S/390.... just got to fire up that S/390 in my basement and get to work on it. Or maybe IBM will give me one. ;-)
AIX....
Linux is not supported and will not run on an RS/6000 S80. I believe Linux us only supported on low-end RS/6000s up to 4-way. The Linux kernel needs lots of work to support a 24-way SMP system.
Good article. But a bit misleading. IBM had a Linux strategy that was announced in March, 1999. This recent announcement builds on that work of 1998/early 1999.
IBM's Linux strategy development got underway in 1998 not Oct 1999.
This is server SW license sales. Which includes preloads, CDs and upgrades to existing installed Linux server SW licenses. Does not include copies or downloads or throw-aways.
Given Linux's business model, one should not expect much revenue from the OS...ever. (ref: Bob Young's goal about collapsing the OS market to $500m from several billions).
What is more interesting is how much new HW (units and revenue) that goes with this. And of course, these unit numbers are great news for SW middleware providers and services companies as it represents a great expansion of opportunity (and no MS and BackOffice to compete with!).
Army No. Va.
I agree with some of the OSS points in the replies here so will not embelish on them. But I do want to make another point.
I'm walking the floor of LinuxWorld and seeing all of the excitement, etc....It's fun, lots of cool stuff, etc... Reminds me of the old USENET and Uniforum conferences in the mid to late 1980s when UNIX was the next big thing.
Yes, some of those companies died or if luckier were bought at a nice price (Apollo comes to mind). Some though went on to become very successful (e.g. Sun, Oracle) *despite* the IBMs and the DECs of the world getting into the UNIX world.
I'm not sure what the lesson OSS leaders are going to learn as implied by this article is, but it certainly appears that they are in a much more upbeat mood and excited and having more fun with cooler SW than say the Windows 2000 development team?!? Even if Win2000 is not a "train wreck" as ESR says, many in the development team must be so burnt out by now and depressed from all the negative press that *they* will feel like they were *in* a train wreck.
The OSS leaders do not have to fear this fate for reasons stated in other replies.
Army of Northern Virginia
Sun does not sell Intel HW, which is what >90% of Linux is based on. Sun does not have a sigifnificant revenue/profit producing SW business nor services business. So, Sun has really no way to make significant money with Linux except perhaps to sell large scale Sun servers as a database server behind many smaller Linux web servers and things like that.
IBM has an Intel HW business and SW and Services.
So, yes, Sun will be on the defensive and will probably have to incrementally change its strategy over time.
Can you say "Let's build Intel servers", Scott?
Army No Va
AIX's origin was System V.1 and Interactive starting in 1983. It first shipped on RT PC in 1986, PS/2 and S/370 in 1988, then RS/6000 in 1990. BSD additions came in the late 80s. AIX was a key foundation for OSF/1, but IBM chose to stay with AIX primarily due to customers not wanting to absorb switching costs.
Solaris was originally BSD based starting in 1981-2. Sometime later, Sun incoporated System V APIs and renamed SunOS to Solaris. Moved to Sparc from Motorola in 1987.
AIX has consistently been a better "enterprise" OS than Solaris as rated by several different consultants. However, Sun has "celebrities", stronger marketing, better visibility and its Internet roots.
Linux will take away much of the low end UNIX server market from both of them due to its business and development models and people with Linux skills coming from school these days.
Army No Va
Several years ago, I was a 2nd line manager in a major UNIX OS organization. Our teams were dissatisfied with the process of working across groups and productivity was low. Along came an ISO9000 "initiative" from above. Rather than grumble about it, we got a team of the top technical leaders as well as some entry people of the organization and let them define how to go about it. The end result: happier and more productive employees, *reduced* overtime and frustration (read....build breaks, breaking more things than fixing, etc...), less stress and gaming for "visibility" or "position", etc...
It's not easy, but it's been done and with benefit. Just have to make sure it does not turn into a beauracratic process, injecting more inefficiency than productivity, though they'll be bumps along the way.