HP on fast track to finding a CEO
JF Mezei
Rumours abound that the choice will be announced by the end of March.
One name mentioned is that of Richard E. Belluzzo -- a longtime HP exec
who now runs data-storage company Quantum.
He spent 23 years at HP already. He had risen to the head of printer and
compuyter businesses. He is a protégé of one of the board members.
##
But some view with skepticism Belluzzo's post-HP track record, which
includes an unsuccessful turnaround effort at Silicon Graphics (SGI )
and a rocky stint as the No. 3 exec at Microsoft (MSFT ). And as CEO of
Quantum since 2002, Belluzzo has seen the company's stock price inch
up 3% during his tenure.
##
So, they'd be hiring a printer guy who worked for Microsoft and wasn't
too succesful at his other jobs.
Not too promising for VMS.
Why can't they find someone of calibre of Gerstner's calibre ?
Bill Todd
...
> Why can't they find someone of calibre of Gerstner's calibre ?
Because HP is a very pale imitation of IBM - even of the failing IBM
that existed when Gerstner took over.
That IBM, while greatly troubled by failed execution, retained most of
its potential. Most of HP's historical potential has been squandered.
- bill
Rob Young
> Not too promising for VMS.
Come on purse your lips... give us a "death of VMS." Just
for old times sake. The good old days of "VMS death", ah... I harken
back:
=============================================================================
From: mezei...@eisner.decus.org (Jean-Francois Mezei)
Date: 1996/02/04
customers who were already fearing the death of VMS.
=============================================================================
From: JF Mezei <jfmezei.spam...@videotron.ca>
Date: Thu, 19 Jul 2001 10:17:04 -0400
current VMS customers would know that VMS's death was impending and switch
vendors
=============================================================================
From: JF Mezei <jfmezei.spam...@videotron.ca>
Date: Fri, 17 May 2002 05:01:22 -0400
VMS's death has been predicted by many, including Palmer et all, and Gartner,
for a long long time.
[ Can't resist - isn't this Generallisimo Francisco
Franco all over again? I mean talk about deja vu:
The death of Spanish dictator Francisco Franco during the first season of
Saturday Night Live in 1975 served as the source of one of the first catch
phrases from SNL to enter the general lexicon.
Franco lingered near death for weeks before dying. On slow news days, United
States network television news casters sometimes noted that Franco was still
alive, or not yet dead. The imminent death of Franco was a headline story on
the NBC news for a number of weeks prior to his death on November 20.
VMS - The Generallisimo Francisco Franco of OSes! ]
=============================================================================
From: JF Mezei <jfmezei.spam...@videotron.ca>
Date: Thu, 22 Nov 2001 11:40:07 -0500
Since it is pretty well a known fact that Compaq or HP will eventually annouce
the death of VMS,
=============================================================================
From: JF Mezei <jfmezei.spam...@vl.videotron.ca>
Date: 2000/03/12
to newsgroups and continue to complain that Compaq is not doing anything to
help VMS escape its death warrant.
=============================================================================
From: jf mezei <"[non-spam]jfmezei"@videotron.ca>
Date: 1997/08/20
(even if DEC doesn't make much money on them) would go a long way to
kill the rumour of VMS's impending death.
=============================================================================
[Need I say: etc. etc.? There is a virtual treasure trove of
"JF death on VMS" favs - come on kids, join the fun! Or act now
and send $9.95 and I'll rush you my top 20 , "JF death on VMS"
You'll laugh, you'll cry... ]
I actually prefer the "impendendings" and the creativity of
"death warrant." Each with their own flavor of weightiness
(as if death itself isn't a weighty subject.) But with "death
warrant" you almost visualize that leftist favorite "Dead Man
Walking" and the touching acting by Penn/Sarandon, et al. Talk
about weight!
Rob
John Smith
> In article <1110417610.15cdca0eb3ae2e5472704358fed95dad@teranews>, JF
> Mezei <jfmezei...@teksavvy.com> writes:
>
>> Not too promising for VMS.
>
> Come on purse your lips... give us a "death of VMS." Just
> for old times sake. The good old days of "VMS death", ah... I harken
> back:
>
>
Been drinking the Kool-Aid again Rob?
--
OpenVMS - The often imitated but never advertised operating system.
Rob Young
>
> Been drinking the Kool-Aid again Rob?
>
Nope. How goes it Johnny Anon?
Rob
pr...@prep.synonet.com
> Why can't they find someone of calibre of Gerstner's calibre ?
Because people of that ability and *credibillity* are very thin
on the ground. The IBM board is one of the best connected `clubs'
on the planet, so anyone who gets their tick is half way home
already.
--
Paul Repacholi 1 Crescent Rd.,
+61 (08) 9257-1001 Kalamunda.
West Australia 6076
comp.os.vms,- The Older, Grumpier Slashdot
Raw, Cooked or Well-done, it's all half baked.
EPIC, The Architecture of the future, always has been, always will be.
John Smith
Don't go blaming JF for the sorry state of the VMS installed base, stuck at
the mythical 411,000 systems for the past, what...8 years or so.
Just treading water and staying at 411,000 - if that's even accurate - is as
good as being dead in the marketplace. It isn't a victory.
--
HP - Where Integrity means "No EV8 for You!!!"
OpenVMS - The often imitated but never advertised operating system.
I could have had an EV8 but all I got was this lousy Itanic instead.
David J Dachtera
>
> In article <1110417610.15cdca0eb3ae2e5472704358fed95dad@teranews>, JF Mezei <jfmezei...@teksavvy.com> writes:
>
> > Not too promising for VMS.
>
> Come on purse your lips... give us a "death of VMS."
Well, let me put it this way: there was another post recently on one of
the mailing lists I subscribe to at work indicating that a VMS shop
would be migrating to AIX for reasons that echo statements made in this
forum. (Such are not uncommon.)
I forwarded it to HP/VMS management with the subject line, "Another
Endangered Account". (Again, not uncommon for me to do that.)
Despite much vaunted - but so far, little documented - growth in the VMS
market, the 'net is rife with evidence of attrition.
Decision makers and juries alike base their decisions on the evidence
presented.
--
David J Dachtera
dba DJE Systems
http://www.djesys.com/
Unofficial OpenVMS Hobbyist Support Page:
http://www.djesys.com/vms/support/
Unofficial Affordable OpenVMS Home Page:
http://www.djesys.com/vms/soho/
Unofficial OpenVMS-IA32 Home Page:
http://www.djesys.com/vms/ia32/
Coming soon:
Unofficial OpenVMS Marketing Home Page
Main, Kerry
David,
Re: moving to AIX ... Course, one when looking at long term futures,
albeit a bit dated, they might also want to consider the following
public statement from the top VP in charge of software at IBM:
http://news.com.com/2100-1001-982512.html?tag=fd_lede2_hed
"NEW YORK--The day is approaching when Linux will likely replace IBM's
version of Unix, the company's top software executive said, an
indication that the upstart operating system's stature is rising within
Big Blue.
While IBM doesn't expect Linux to replace its own AIX version of Unix
any time soon, Big Blue is pushing the open-source OS in the that
direction, Steve Mills, senior vice president of IBM's Software Group,
told CNET News.com at last week's LinuxWorld trade show.
Asked whether IBM's eventual goal is to replace AIX with Linux, Mills
responded, "It's fairly obvious we're fine with that idea...It's the
logical successor."
Regards
Kerry Main
Senior Consultant
HP Services Canada
Voice: 613-592-4660
Fax: 613-591-4477
kerryDOTmainAThpDOTcom
(remove the DOT's and AT)
"OpenVMS has always had integrity ..
Now, Integrity has OpenVMS .."
Brad Hamilton
<snip>
> David,
>
> Re: moving to AIX ... Course, one when looking at long term futures,
> albeit a bit dated, they might also want to consider the following
> public statement from the top VP in charge of software at IBM:
>
> http://news.com.com/2100-1001-982512.html?tag=fd_lede2_hed
<snip>
Kerry,
That article is now more than two years old - where's the death? :-)
David is (I think) speaking of a vendor whose product used to be
available exclusively on VMS, but is now available on AIX, as well. I
was acquainted with a hospital in my area that started testing one of
these boxes last year (code named: Shark - how's _that_ for irony?)
There was a VMS box in the same computer room, running a different
application, but the handwriting was plainly on the wall for me - "old"
VMS box (2100 - V6.2-1H1) on the way out, "new" AIX box(es) (about the
size and shape of the EV Storageworks boxes) on the way in.
HP (VMS) is rapidly losing ground in one of its last "niche" markets,
and prefers to stand by, idly whistling in the dark, and hoping its
competitors will just go away quietly.
JF Mezei
> That article is now more than two years old - where's the death? :-)
>
> David is (I think) speaking of a vendor whose product used to be
> available exclusively on VMS, but is now available on AIX, as well.
We, as VMS supporters, cannot use AIX' future as argument because VMS is
exactly in the same shoe, except that AIX is onwed and supported by a
reputable vendor, VMS isn't.
The big difference is that with the backing of IBM, when AIX is retired,
you can bet your bottom that your strategic applications will run on
Linux and IBM will be there to help you and IBM will come thorugh with
the port of any AIX specific features to Linux. (something wich HP has
now failed and can n longer be trusted, after its decision to canecl the
porting of Trucluster to HP-UX).
AIX has a long term path: Linux on Power.
VMS doesn't have a clear future. The OS may have a roadmap of a year or
two, but nobody trusts the vendor's promises on the platform's future
(or lack thereof), and everyone knows that the vendor isn't really
working hard to protect and grow the real applications available on VMS.
And for what it's worth, HP-UX is in a worse boat than AIX. At least AIX
users can migrate to Linux on Power with same endiannes. HP-UX users on
Pa-Risc will have to migrate to either Power, Sparc or 8086 with
different endianness.
Remember that Intel has now admitted that IA64's niche has been further
reduced to just the very high end (the Cray of the 1980s) and that just
isn't ve relevant to common business needs.
DeanW
> -----Original Message----- From: David J Dachtera
> [mailto:djesys...@comcast.net]
> Sent: March 10, 2005 9:53 PM
>
>> one of the mailing lists I subscribe to at work indicating that a
>> VMS shop would be migrating to AIX for reasons that echo statements
>> made in this forum. (Such are not uncommon.)
>>
>> I forwarded it to HP/VMS management with the subject line, "Another
>> Endangered Account". (Again, not uncommon for me to do that.)
>>
>> Despite much vaunted - but so far, little documented - growth in
>> the VMS market, the 'net is rife with evidence of attrition.
>>
>> Decision makers and juries alike base their decisions on the
>> evidence presented.
>
>
> Re: moving to AIX ... Course, one when looking at long term futures,
> albeit a bit dated, they might also want to consider the following
> public statement from the top VP in charge of software at IBM:
With all due respect, Kerry, sometimes I think you (personally, and HP
in general) JUST DON'T GET IT. I've heard YOU, Kerry, blow-off and
hand wave these things personally, plus the announcements (or lack of)
from HP indicate that they're not paying attention. What was the last
one- that LSE or someone was going to Windows? "Oh, they'll be sorry
about that!" someone [That was Kerry again, I think] said. Sure,
they'll be sorry- but THEY'LL BE ON WINDOWS, FOR GOOD OR BAD. Hello-
does that sound like "not a VMS customer"? Having left, maybe they'll
come back- but given the cause for the LSE move (VMS is too
expensive), probably they'll hunker down and make the MS stuff work
acceptably (FSVO), or they'll try something else. Big decision makers
don't like to go back- that looks too much like they were wrong, and
being wrong is bad for a career.
The issue is- this customer is thinking about going away from VMS,
based on perceptions (VMS is too expensive to own, or develop for) and
maybe because of FUD- and all HP does is try to FUD back- with rather
anemic FUD, at that. What can you trot out to counter the constant
"VMS is going away" messages that we hear everywhere? The "VMS is
growing" is vague and appears unsupported in face of anecdotal
evidence and a recent history of [percieved] misleading statements or
broken promises. Whether those statements were misleading, or promises
actually were broken, I'll leave for others to decide- the
_perception_ is that they were, and it's the _perception_ that counts.
There's only one company that's going to debunk those- don't expect
IBM or Sun to say "Oh, we're sure VMS will be around for a long while!
And it's TCO and development costs are lower than our offerings, too.
Would you like some AIX now, or should I hook you up with an HP rep?"
OK, so IBM says Linux may one day replace AIX. When that happens, you
can bet that IBM will make that transition as perfectly transparent
and painless as possible- quite likely very transparent and painless-
IBM is spending quite a chunk on Linux development. And their
investment in hardware is safe- Linux runs on damn near anything. So,
to counteract the "AIX is going away" FUD, the IBM rep says "Sure.
AIX, Linux- pick one. They both run on Power, or you can get Linux on
x86 if you'd like. In a few years all those features will be merged
together and IBM can have you covered for servers small, medium, or
large, all on one OS. So, what flavor would you like to start with?"
Someone from HP needs to run down the customer Dave mentioned, find
out what the _real_ issues are (just like when someone asks for help
here), and help them solve them. If that happens to be an HP VMS
ambassador, great. Otherwise, the name on that customer's next box
will probably be "IBM", because nobody's given them a reason why it
shouldn't be.
Bob Kaplow
> the port of any AIX specific features to Linux. (something wich HP has
> now failed and can n longer be trusted, after its decision to canecl the
> porting of Trucluster to HP-UX).
Don't forget the previous commitment to port Tru64 to IA64...
Bob Kaplow NAR # 18L TRA # "Impeach the TRA BoD"
>>> To reply, remove the TRABoD! <<<
Kaplow Klips & Baffle: http://nira-rocketry.org/LeadingEdge/Phantom4000.pdf
www.encompasserve.org/~kaplow_r/ www.nira-rocketry.org www.nar.org
They that can give up essential liberty to obtain a little
temporary safety deserve neither liberty nor safety. --
Benjamin Franklin Historical Review of Pennsylvania. 1759
Main, Kerry
> From: JF Mezei [mailto:jfmezei...@teksavvy.com]
> Sent: March 11, 2005 11:01 AM
> To: Info...@Mvb.Saic.Com
> Subject: Re: HP on fast track to finding a CEO
>
> > That article is now more than two years old - where's the
> death? :-)
> >
> > David is (I think) speaking of a vendor whose product used to be
> > available exclusively on VMS, but is now available on AIX, as well.
>
> We, as VMS supporters, cannot use AIX' future as argument
> because VMS is
> exactly in the same shoe, except that AIX is onwed and supported by a
> reputable vendor, VMS isn't.
>
> The big difference is that with the backing of IBM, when AIX
> is retired,
> you can bet your bottom that your strategic applications will run on
> Linux and IBM will be there to help you and IBM will come thorugh with
> now failed and can n longer be trusted, after its decision to
> canecl the
> porting of Trucluster to HP-UX).
For things on Linux that require kernel changes, IBM can only request
features. Linus (and associates) are the only ones who decide what and
*when* kernel changes get done.
>
> AIX has a long term path: Linux on Power.
So why should a Customer move all their stuff from OpenVMS (especially
if they are active-active clustering) to an active-passive fail-over
environment on AIX and then move *again* (with all of the porting,
testing etc) to Linux?
If this is the direction the Customer wants to go, why not go directly
to Linux on Power?
> VMS doesn't have a clear future. The OS may have a roadmap of
> a year or
> two, but nobody trusts the vendor's promises on the platform's future
> (or lack thereof), and everyone knows that the vendor isn't really
> working hard to protect and grow the real applications
> available on VMS.
>
I have no doubt AIX will be supported for a long time, but since you
mentioned the roadmap, show me where you can publicly get a long term
roadmap for AIX futures on the IBM web site. This is not a jab - I
really would like to see what they are publicly announcing for AIX
futures - and how many years out the roadmap goes.
> And for what it's worth, HP-UX is in a worse boat than AIX.
> At least AIX
> users can migrate to Linux on Power with same endiannes.
> HP-UX users on
> Pa-Risc will have to migrate to either Power, Sparc or 8086 with
> different endianness.
>
>
> Remember that Intel has now admitted that IA64's niche has
> been further
> reduced to just the very high end (the Cray of the 1980s) and
> that just
> isn't ve relevant to common business needs.
So, are you saying business requirements no longer include RISC i.e.
server systems - the target market for Itanium?
And deliberately changing what Intel says to map to your own agenda is
not helpful. While you may not agree with them, Intel directions are
summarized in the following:
http://news.zdnet.com/2100-9584_22-5593047.html (Feb 28, 2005)
Intel to spotlight new Itanium: 'Poulson'
[snip...]
"But for all the difficulties, Intel isn't throwing in the towel.
One major change will come with the Tukwila generation of chips. At that
point, for the first time, Xeon and Itanium chips will have the same
electronic interface, making it easier to design servers that support
either processor. And Intel promises that in 2007, Itanium and Xeon
systems will cost the same, but Itaniums will have twice the
performance."
Bill Gunshannon
"Main, Kerry" <kerry...@hp.com> writes:
>> -----Original Message-----
>> From: JF Mezei [mailto:jfmezei...@teksavvy.com]=20
>> Sent: March 11, 2005 11:01 AM
>> To: Info...@Mvb.Saic.Com
>> Subject: Re: HP on fast track to finding a CEO
>>=20
>> Brad Hamilton wrote:
>> > That article is now more than two years old - where's the=20
>> death? :-)
>> >=20
>> > David is (I think) speaking of a vendor whose product used to be
>>=20
>> We, as VMS supporters, cannot use AIX' future as argument=20
>> because VMS is
>> exactly in the same shoe, except that AIX is onwed and supported by a
>> reputable vendor, VMS isn't.
>> The big difference is that with the backing of IBM, when AIX=20
>> is retired,
>> you can bet your bottom that your strategic applications will run on
>> Linux and IBM will be there to help you and IBM will come thorugh with
>> the port of any AIX specific features to Linux. (something wich HP has
>> canecl the
>> porting of Trucluster to HP-UX).
>
> For things on Linux that require kernel changes, IBM can only request
> features. Linus (and associates) are the only ones who decide what and
Actually, that's not true. There is nothing stopping IBM from spawning
their own distribution with their own series of kernels. The only
requiremetn of the Gnu Public Virus is that they make anuy of their
own modifications publicly available. It would be interesting to see
how long it would take, in such an event, before IBM's kernel was the
defacto standard.
Of course, I still can't imagine why they (or any business) use
Linux at all instead of one of the available BSD's with a much
more commercial friendly license.
bill
--
Bill Gunshannon | de-moc-ra-cy (di mok' ra see) n. Three wolves
bi...@cs.scranton.edu | and a sheep voting on what's for dinner.
University of Scranton |
Scranton, Pennsylvania | #include <std.disclaimer.h>
Dave Froble
Support for Linux by IBM isn't a trivial thing. If IBM really needs
something, and can justify it, I think that there will be ears willing
to listen. Hey, if it's not a 2-way street, then it's a street to nowhere.
>>AIX has a long term path: Linux on Power.
>
>
> So why should a Customer move all their stuff from OpenVMS (especially
> if they are active-active clustering) to an active-passive fail-over
> environment on AIX and then move *again* (with all of the porting,
> testing etc) to Linux?
>
> If this is the direction the Customer wants to go, why not go directly
> to Linux on Power?
Moving without strong reasons is just plain stupid. That said, I've
seen such moves in the past that weren't in the interest of the company,
but in the interest of the employees, who were able to convince the
company to go along. There has been more than one instance of someone
claiming that VMS wasn't good for their career. It happens.
>>VMS doesn't have a clear future. The OS may have a roadmap of
>>a year or
>>two, but nobody trusts the vendor's promises on the platform's future
>>(or lack thereof), and everyone knows that the vendor isn't really
>>working hard to protect and grow the real applications
>>available on VMS.
>>
>
>
> I have no doubt AIX will be supported for a long time, but since you
> mentioned the roadmap, show me where you can publicly get a long term
> roadmap for AIX futures on the IBM web site. This is not a jab - I
> really would like to see what they are publicly announcing for AIX
> futures - and how many years out the roadmap goes.
Good shot Kerry. Everyone makes much over the VMS roadmap, but VMS may
have one of the most extensive roadmaps. Only problem is it's hidden.
Yeah, I know it's available on the web site, but that's only good if
someone knows to go to the web site. Back to marketing.
If itanic actually earns such a performance gap, then that will be good
for those depending upon itanic. However, Intel cannot try to provide
the gap by holding IA-32/x86 back. They already got their ass's kicked
by AMD once, they won't make that mistake again.
Main, Kerry
> From: Dave Froble [mailto:da...@tsoft-inc.com]
> Sent: March 11, 2005 5:27 PM
> To: Info...@Mvb.Saic.Com
> Subject: Re: HP on fast track to finding a CEO
>
[snip ...]
> > If this is the direction the Customer wants to go, why not
> go directly
> > to Linux on Power?
>
> Moving without strong reasons is just plain stupid. That said, I've
> seen such moves in the past that weren't in the interest of
> the company,
> but in the interest of the employees, who were able to convince the
> company to go along. There has been more than one instance
> of someone
> claiming that VMS wasn't good for their career. It happens.
>
Yep, especially when staff includes strong contingent of external
consultants who are very product focussed. Recently seen example where
a new application requirements came out with "must support Oracle 10G".
Not must have [insert latest industry SQL stds, transaction levels, mgmt
capabilities etc.], but very specific product reference. And this was
for environment with no previous production implementation of Oracle.
Of course, those consultants on staff were primarily private Oracle
consulting backgrounds who wanted the latest and greatest on their
resumes.
[snip..]
pr...@prep.synonet.com
> For things on Linux that require kernel changes, IBM can only
> request features. Linus (and associates) are the only ones who
> decide what and *when* kernel changes get done.
Horse stuff...
They have the code, there is NOTHING to stop them from writing it
themselves! Zip, nadda, nil. But they DO have to give the source
to their customers.
What other parts of Open Souce don't you understand?
Karsten Nyblad
> "Main, Kerry" <kerry...@hp.com> writes:
>
>
>>For things on Linux that require kernel changes, IBM can only
>>request features. Linus (and associates) are the only ones who
>>decide what and *when* kernel changes get done.
>
>
> Horse stuff...
>
> They have the code, there is NOTHING to stop them from writing it
> themselves! Zip, nadda, nil. But they DO have to give the source
> to their customers.
>
> What other parts of Open Souce don't you understand?
>
Kerry is right, but IBM's options look much better if the wording is
changed a little.
Of course IBM can make the modifications that minimizes that work of
moving from AIX to Linux, but the customers will want to run Redhat or
SuSE. Thus IBM will have a strong interest in getting the Linux
community to accept their changes.
Alternatively IBM could make a runtime environment, that made Linux look
like AIX to programs. That could be runtime libraries, a virtual Power
machine, or any of the other tricks that have been used by, e.g.,
Digital to make VAX code run on Alpha, various companies and OSS people
to make Windows code run on none Windows platforms, etc.
Please note that there are environments for making Windows code run on
Linux, and that people have been writing code to run on multiple *nix
platforms for years. It seems feasible to make Linux look like AIX to
programs without changing that much of Linux.
Bill Gunshannon
Karsten Nyblad <nos...@nospam.nospam> writes:
> pr...@prep.synonet.com wrote:
>> "Main, Kerry" <kerry...@hp.com> writes:
>>
>>
>>>For things on Linux that require kernel changes, IBM can only
>>>request features. Linus (and associates) are the only ones who
>>>decide what and *when* kernel changes get done.
>>
>>
>> Horse stuff...
>>
>> They have the code, there is NOTHING to stop them from writing it
>> themselves! Zip, nadda, nil. But they DO have to give the source
>> to their customers.
>>
>> What other parts of Open Souce don't you understand?
>>
> Well, Kerry is spinning the truth to make VMS & HP look good. I think
> Kerry is right, but IBM's options look much better if the wording is
> changed a little.
>
> Of course IBM can make the modifications that minimizes that work of
> moving from AIX to Linux, but the customers will want to run Redhat or
> SuSE. Thus IBM will have a strong interest in getting the Linux
> community to accept their changes.
If they put their minds to it, do you think it would be all that long
before IBM supplanted RedHat and/or SuSE as the ruler of the commercial
Linux world? Even Linus has nothing to protect his position as the
God of Linux.
>
> Alternatively IBM could make a runtime environment, that made Linux look
> like AIX to programs. That could be runtime libraries, a virtual Power
> machine, or any of the other tricks that have been used by, e.g.,
> Digital to make VAX code run on Alpha, various companies and OSS people
> to make Windows code run on none Windows platforms, etc.
>
> Please note that there are environments for making Windows code run on
> Linux, and that people have been writing code to run on multiple *nix
> platforms for years. It seems feasible to make Linux look like AIX to
> programs without changing that much of Linux.
Assuming the availbility of AIX Libraries (and I assume IBM has that :-)
there would be very little involved in making AIX software run on any
of the Open Source systems. Linux code runs on BSD using this very method.
All it is going to take is for IBM to finally get serious about this sector
of the market.
Karsten Nyblad
> In article <42333034$0$78287$157c...@dreader1.cybercity.dk>,
> Karsten Nyblad <nos...@nospam.nospam> writes:
>>Of course IBM can make the modifications that minimizes that work of
>>moving from AIX to Linux, but the customers will want to run Redhat or
>>SuSE. Thus IBM will have a strong interest in getting the Linux
>>community to accept their changes.
>
>
> If they put their minds to it, do you think it would be all that long
> before IBM supplanted RedHat and/or SuSE as the ruler of the commercial
> Linux world? Even Linus has nothing to protect his position as the
> God of Linux.
Right now you need support for Oracle databases if you want to be a
Linux vendor. Oracle would be dead dumb if they helped their largest
competitor on DB software getting a monopoly on the Linux platform.
IBM has been spending much money on Linux, but it is difficult to see
that they have got that much revenue from that. Linux fans may like
what IBM do, but the fans do not let that decide their purchase
decisions. Sun have written important OSS like OpenOffice and J2EE is
the only competitor to .NET, but Sun is in trouble. IBM might be
writing great software for Linux, but I doubt they can wind that many
customers that way.
And now my own point of view. I am a geek that hates monopolies, one
that likes the idea of software being free as in speech, not only free
as in beer.
The VMS zealots always points out that closed source software has the
advantage that if you have a problem, you can go to the software vendor
and make the vendor fix it. That might be true if you are a big
customer of the vendor. Otherwise you can send in bug reports, and if
you are lucky, you get a fix 6-12 months later. If you are a small
customer, than you are in a constant risk of running into a problem that
is a major issue to you, but not to other customers, and the vendor not
willing to fix the problem. Face it: The support model of open source
software has problems, but so has the support model of closed source. I
can't see that the support model of closed source is any better to small
customer than the support model of OSS.
Then you have the advantage that you can call the vendor if you have
some dumb questions on how to use their software. Well I have only
tried it with VMS, and the Danish support could only answer questions
were you could read the answer in the manuals plus questions on how to
get the system parameters right. Not that much help to me. I would say
that reading the manuals plus c.o.v would put most people i a better
situation than using the Danish support.
OK, documentation of OSS software sucks and so does the quality of the
less poppular programs, but you can look into the source and in most
cases it is not that difficult to get your questions answered that way.
And most software bugs can be fixed or circumvented with minor changes
to the programs.
Then there is Windows. From my point of view as an IT person that
platform really combine the problems of closed source software with the
problems of open source software. If you want real support, then you
have to pay real money. Otherwise you are stuck with a poorly
documented OS, were you can use very long time to troubleshoot simple
problems simply because you do not know what is going on inside Windows.
Further, software for software development is NOT free for any of the
closed source platforms. E.g. on Windows you have to subscribe to MSDN
and buy Visual Studio and a database if you what to do any serious
programming.
I do not want to go back to a model, where a single vendor controls the
platform. Before Microsoft became the hate object of everybody not
liking monopolies it was IBM that had status. I would prefer that IBM
does not get that position again, and thus I would be tempted to buy
from other vendors should IBM try to monopolies Linux.
JF Mezei
> IBM has been spending much money on Linux, but it is difficult to see
> that they have got that much revenue from that.
IBM's Linux moves are far more strategic than it appears on the surface.
You need to remmener where IBM started from, a company viewed as a
dinosaur with its mainframes and made irrelevant in the IT industry.
From a PR perspective, the very visible Linux support helps position IBM
as a leader in the industry. (as opposed to an also-ran).
Being seen as a leader also helps push Power based machines to host
Linux. It also tried this with its 370 mainframes earlier on.
Beeing seen as a leader also helps IBM services gets contracts to deploy
and support Linux based systems. Linux may be free, but support isn't.
Another point is with IBM's own installed base. If IBM weren't seen as
linux expert, customers might not consider IBM and go with other vendors
for new Linux projects, which woudl result in net loss for IBM. With a
strong image of Linux expertise and solutions, IBM customers will find
it very easy to stay with IBM should they consider moving from
MVS/AIX/AS400 to Linux.
Tom Linden
> Karsten Nyblad wrote:
>> IBM has been spending much money on Linux, but it is difficult to see
>> that they have got that much revenue from that.
>
>
> IBM's Linux moves are far more strategic than it appears on the surface.
> You need to remmener where IBM started from, a company viewed as a
> dinosaur with its mainframes and made irrelevant in the IT industry.
IBM has experienced a renaissance with the mainframes. Linux will run in
an LPAR, for example.
>
> From a PR perspective, the very visible Linux support helps position IBM
> as a leader in the industry. (as opposed to an also-ran).
>
> Being seen as a leader also helps push Power based machines to host
> Linux. It also tried this with its 370 mainframes earlier on.
>
> Beeing seen as a leader also helps IBM services gets contracts to deploy
> and support Linux based systems. Linux may be free, but support isn't.
>
> Another point is with IBM's own installed base. If IBM weren't seen as
> linux expert, customers might not consider IBM and go with other vendors
> for new Linux projects, which woudl result in net loss for IBM. With a
> strong image of Linux expertise and solutions, IBM customers will find
> it very easy to stay with IBM should they consider moving from
> MVS/AIX/AS400 to Linux.
I get the impression that IBM and others are involved with Linux because
they
don't want anybody else to run away with the business, should it prove to
be
viable.
Bill Gunshannon
Karsten Nyblad <nos...@nospam.nospam> writes:
> Bill Gunshannon wrote:
>> In article <42333034$0$78287$157c...@dreader1.cybercity.dk>,
>> Karsten Nyblad <nos...@nospam.nospam> writes:
>>>Of course IBM can make the modifications that minimizes that work of
>>>moving from AIX to Linux, but the customers will want to run Redhat or
>>>SuSE. Thus IBM will have a strong interest in getting the Linux
>>>community to accept their changes.
>>
>>
>> If they put their minds to it, do you think it would be all that long
>> before IBM supplanted RedHat and/or SuSE as the ruler of the commercial
>> Linux world? Even Linus has nothing to protect his position as the
>> God of Linux.
>
> Right now you need support for Oracle databases if you want to be a
> Linux vendor. Oracle would be dead dumb if they helped their largest
> competitor on DB software getting a monopoly on the Linux platform.
Well, if IBM is pushing Linux in any flavor and Oracle is supporting
Linux it seems they are already helping their largest competitor.
Fact is that IBM may be a competitior but they wouldn't hesitate to
sell Oracle if that's what the customer wanted.
>
> IBM has been spending much money on Linux, but it is difficult to see
> that they have got that much revenue from that.
And what revenue does RedHat get from Linux? One would assume pretty
much all of it. And IBM could expect the same if they offered the
same services that RedHat did. I would be willing to bet (although
I am sure there is no more chance of verifying it than verifying the
same for VMS) that IBM is making money from Linux. They didn't get
where they are by being philanthropists.
> Linux fans may like
> what IBM do, but the fans do not let that decide their purchase
> decisions. Sun have written important OSS like OpenOffice and J2EE is
> the only competitor to .NET, but Sun is in trouble. IBM might be
> writing great software for Linux, but I doubt they can wind that many
> customers that way.
Your right. RedHat is where they are because of the package and
services they offer. IBM is not only in a position to offer more,
they are also in a position to blow their own horn even louder
than RedHat. And when you figure that the biggest thing Linux has
going for it is the hype the name IBM is more than enough to take
over the market if they ever decide they care.
JF Mezei
> I get the impression that IBM and others are involved with Linux because
> they don't want anybody else to run away with the business, should it prove to
> be viable.
I woudl generally agree with this assesment. However, I think IBM not
only doesn't want to miss the boat, but also wan't to be in the
wheelhouse, directing where the boat goes and be seen as a leader in
that market. And IBM is using the leadership it has regained under
Gerstner to make significant push in teh Linux market.
Companies such as HP who are still loyal to Microsoft, are selling Linux
just so that they don't miss the boat. They aren't pushing Linux against mcirosoft.
JF Mezei
> Well, if IBM is pushing Linux in any flavor and Oracle is supporting
> Linux it seems they are already helping their largest competitor.
> Fact is that IBM may be a competitior but they wouldn't hesitate to
> sell Oracle if that's what the customer wanted.
Gertsner talks about this problematic issue in his book. He did make the
convincing argument to sell certain products to 3rd parties in order to
not be seen competing against a more popular vendor. But IBM also made
key decisions to keep on competing in the database market with its DB2.
It took a lot of cultural change at IBM to get the sales reps to start
to work with competitors to build proposals that involved both IBM and
competitor's products. (I think that the term that was coined was
coopetition) (cooperation + competition).
DEC could have kept RDB if its sales reps were more opened to selling
Oracle and let cistomers decide between RDB and Oracle.
> Your right. RedHat is where they are because of the package and
> services they offer. IBM is not only in a position to offer more,
> they are also in a position to blow their own horn even louder
> than RedHat.
But the reason that IBM is able to blow their horn about Linux is very
important: IBM regained a leadership position in the market and has
become vbery relevant, thanks to the changes made by Gerstner and the
MARKETING that was done to bring IBM back into the 1990s and jumpstard a
sick old mainframe proprietary expensive company into a leader for
internet software/services for enterprise.
Where does HP really lead in the enterprise sector ? Not Linux, not
Unix, not the chip (Power is better than IA64). They certaintly don't
have much credibility with SAP.
HP does a lot of things "acceptably". But they don't seem to excell in
any of them. They just follow want Intel and Microsoft tell them to do.
In fact, from a practical point of view, HP appears to be more of a
joint venture between Microsoft and Intel.
MS wants to go to consumer goods and home entertainment ? HP goes
consumer goods and home entertainment.
The vendors should be telling Intel and Microsoft what to do, not the
other way around. It is the ven
dors who have the contact with the customers. IBM did a very smart thing
by distancing itseof from microsoft when it realised that a relationship
with Microsoft is always one of slave, with the vendor nevert having any
ifluence over microsoft. By being independant, you can direct your
products faster towards what your customers need than if you just follow
what microsoft decides to do, especially in the enterprise business.
Dave Froble
>
> I get the impression that IBM and others are involved with Linux
> because they
> don't want anybody else to run away with the business, should it prove
> to be
> viable.
>
I seem to remember the advice on how to be a leader. Find out which way
the mob is headed, get to the front, and shout "Follow me!"
Main, Kerry
> -----Original Message-----
> From: pr...@prep.synonet.com [mailto:pr...@prep.synonet.com]
> Sent: March 12, 2005 10:38 AM
> To: Info...@Mvb.Saic.Com
> Subject: Re: HP on fast track to finding a CEO
>
>
> > For things on Linux that require kernel changes, IBM can only
> > request features. Linus (and associates) are the only ones who
> > decide what and *when* kernel changes get done.
>
> Horse stuff...
>
> They have the code, there is NOTHING to stop them from writing it
> themselves! Zip, nadda, nil. But they DO have to give the source
> to their customers.
>
> What other parts of Open Souce don't you understand?
>
Great - another UNIX kernel distribution for all of the ISV's Like
Oracle to support. ISV's just love having more platforms to add to all
of their support challenges.
:-)
Remember the old "it's the applications - not the OS" argument so
familiar on this newsgroup?
Do you really think IBM is going to go up against Red Hat and SuSe with
another Linux kernel distribution that competes with theirs? What do you
think Red Hat and Suse would do with regards to their Power platform
support if IBM came up with another Linux pkg?
Paul Sture
> On Sun, 13 Mar 2005 01:50:03 -0500, JF Mezei <jfm...@teksavvy.com> wrote:
>
>> Karsten Nyblad wrote:
>>
>>> IBM has been spending much money on Linux, but it is difficult to see
>>> that they have got that much revenue from that.
>>
>>
>>
>> IBM's Linux moves are far more strategic than it appears on the surface.
>> You need to remmener where IBM started from, a company viewed as a
>> dinosaur with its mainframes and made irrelevant in the IT industry.
>
>
> IBM has experienced a renaissance with the mainframes. Linux will run in
> an LPAR, for example.
>
I have heard of mainframes running a large number of Linux instances.
Existing mainframe owners can test such configurations without buying a
fleet of linux boxes for an evaluation; others may wish to consolidate
an existing fleet onto their mainframe.
Of course IBM will be more than happy to sell extra processing power
should it be required, and that is where I see them making money.
Also, isn't it true that the supported versions of Linux (SuSE, Red Hat)
don't come free, as you have to buy support licences?
>>
>> From a PR perspective, the very visible Linux support helps position IBM
>> as a leader in the industry. (as opposed to an also-ran).
>>
>> Being seen as a leader also helps push Power based machines to host
>> Linux. It also tried this with its 370 mainframes earlier on.
>>
>> Beeing seen as a leader also helps IBM services gets contracts to deploy
>> and support Linux based systems. Linux may be free, but support isn't.
>>
>> Another point is with IBM's own installed base. If IBM weren't seen as
>> linux expert, customers might not consider IBM and go with other vendors
>> for new Linux projects, which woudl result in net loss for IBM. With a
>> strong image of Linux expertise and solutions, IBM customers will find
>> it very easy to stay with IBM should they consider moving from
>> MVS/AIX/AS400 to Linux.
>
>
> I get the impression that IBM and others are involved with Linux
> because they
> don't want anybody else to run away with the business, should it prove
> to be
> viable.
>
But IBM can make money from selling the resources and support for Linux
on mainframes.
Karsten Nyblad
> Do you really think IBM is going to go up against Red Hat and SuSe with
> another Linux kernel distribution that competes with theirs? What do you
> think Red Hat and Suse would do with regards to their Power platform
> support if IBM came up with another Linux pkg?
Since IBM already has an investment in Novell and Novell owns SuSE it
would be more likely that IBM bought Novell or SuSE. But I think,
cusotmers not wanting a new monopoly would leave SuSE, and thus it would
be a poor business decision to IBM.
Karsten Nyblad
> Also, isn't it true that the supported versions of Linux (SuSE, Red Hat)
> don't come free, as you have to buy support licences?
RedHat has put copyrighted logos on their supported version, and you are
only allowed to have these logos on machines with support contracts.
Thus, you cannot install RedHat on machines without support contracts,
and if you cancel the contract, then you have to remove the logos, and
RedHat has not documented how to do that. Thus, in a way RedHat has
less favorable license terms on their Linux than, e.g., bought OpenVMS.
No one forces you to remove OpenVMS if you cancel your support contract.
There are distributions out there, which differ from RedHat only in the
logos.
Dr. Dweeb
Well, to take an example, OracleRdb support seems to turn around fixes quite
pronto, and it seems that the *significance* of the fix required is the
determining factor, rather than the size of the customer who raised the bug
report. I have personally seen bugfixes turned around within hours. My
point is, that just because some organisations (and MS is really guilty
here) see support as a contest to deliver the least amount of information in
the greatest possible time, it is not true of all organisations and long
turnaround time for fixes is a question of policy from the company - not a
problem endemic to proprietary software per se.
> Then you have the advantage that you can call the vendor if you have
> some dumb questions on how to use their software. Well I have only
> tried it with VMS, and the Danish support could only answer questions
> were you could read the answer in the manuals plus questions on how to
> get the system parameters right. Not that much help to me. I would
> say that reading the manuals plus c.o.v would put most people i a
> better situation than using the Danish support.
>
Danish support used to have a number of talented and *very* knowledgable
people. The destruction of Digital from within and the subsequent CompaQ
takeover more or less ended what little there was of VMS in danmark. I
guess support got decimated along with everything else, so I am unsurprised
that your experience with VMS support in danmark is negative. When was the
last significant new VMS client sale in danmark ? I suggest it was over a
decade ago, and that says all you need to know about how much Digital
danmark and since then CompaQ cared about VMS. I doubt that a cold call to
HP for information about VMS would be met with anything other than a
mystified silence.
Many of the talented people in danmark have been forced to either (a) leave
the country to find work, (b) abandon VMS as a career, (c) retire or be
unemployed, or as in my case, all of the above in that sequence. There are
unlikely to be many new people trained, and except for a small number of
important "old" customers, the market is rapidly dwindling to zero. Danmark
might actually be in the running for the first VMS-free country in Europe.
It can therefore only be a matter of time before danish local support is
completely abandoned and outsourced to India.
> OK, documentation of OSS software sucks and so does the quality of the
> less poppular programs, but you can look into the source and in most
> cases it is not that difficult to get your questions answered that
> way. And most software bugs can be fixed or circumvented with minor
> changes to the programs.
>
> Then there is Windows. From my point of view as an IT person that
> platform really combine the problems of closed source software with
> the problems of open source software. If you want real support, then
> you have to pay real money. Otherwise you are stuck with a poorly
> documented OS, were you can use very long time to troubleshoot simple
> problems simply because you do not know what is going on inside
> Windows.
>
Even when you pay "real-money" it does not help that much. When MS screwed
with things seriously between NT4 and XP, and broke things, they simply
never bothered to fix them. It was possible to spend months getting MS top
even acknowledge that a repeatable and different behaviour was even a bug.
Customers were required to change/fix there own installations, sometimes at
significant cost or data loss. That is the power of a monopoly.
> Further, software for software development is NOT free for any of the
> closed source platforms. E.g. on Windows you have to subscribe to
> MSDN and buy Visual Studio and a database if you what to do any
> serious programming.
>
> I do not want to go back to a model, where a single vendor controls
> the platform. Before Microsoft became the hate object of everybody
> not liking monopolies it was IBM that had status. I would prefer
> that IBM does not get that position again, and thus I would be
> tempted to buy from other vendors should IBM try to monopolies Linux.
Gee, are you old enough to remember that. They were certainly high on my
list back when I was programming for S/34 og VSE on the 4381.
Dr. Dweeb
pr...@prep.synonet.com
>> -----Original Message-----
>> From: pr...@prep.synonet.com [mailto:pr...@prep.synonet.com]
>> Sent: March 12, 2005 10:38 AM
>> To: Info...@Mvb.Saic.Com
>> Subject: Re: HP on fast track to finding a CEO
>>
>> "Main, Kerry" <kerry...@hp.com> writes:
>>
>> > For things on Linux that require kernel changes, IBM can only
>> > request features. Linus (and associates) are the only ones who
>> > decide what and *when* kernel changes get done.
>>
>> Horse stuff...
>>
>> They have the code, there is NOTHING to stop them from writing it
>> themselves! Zip, nadda, nil. But they DO have to give the source
>> to their customers.
>>
>> What other parts of Open Souce don't you understand?
> Great - another UNIX kernel distribution for all of the ISV's Like
> Oracle to support. ISV's just love having more platforms to add to
> all of their support challenges.
If IBM offered a near current linux with a libAIX and tools so you
could compile you code with NO changes, you would be killed in the
rush.
> :-)
> Remember the old "it's the applications - not the OS" argument so
> familiar on this newsgroup?
Pity someone does not try to get some. Oh yes, but they have Integrity...
> Do you really think IBM is going to go up against Red Hat and SuSe
> with another Linux kernel distribution that competes with theirs?
> What do you think Red Hat and Suse would do with regards to their
> Power platform support if IBM came up with another Linux pkg?
What CAN RH and Suze do? Please peeeple, IBM are being nasty to
us so we want you all to burn you PPCs and macs...
--
Paul Repacholi 1 Crescent Rd.,
+61 (08) 9257-1001 Kalamunda.
West Australia 6076
DeanW
> Bill Gunshannon wrote:
> > In article <42333034$0$78287$157c...@dreader1.cybercity.dk>,
> > Karsten Nyblad <nos...@nospam.nospam> writes:
> >>Of course IBM can make the modifications that minimizes that work of
> >>moving from AIX to Linux, but the customers will want to run Redhat or
> >>SuSE. Thus IBM will have a strong interest in getting the Linux
> >>community to accept their changes.
> >
> > If they put their minds to it, do you think it would be all that long
> > before IBM supplanted RedHat and/or SuSE as the ruler of the commercial
> > Linux world? Even Linus has nothing to protect his position as the
> > God of Linux.
What makes you think IBM isn't taking the slow sneaky way to doing
that? Up here around Portland, OR, there are a fair number of IBM
engineers working on Linux full-time. Linus has been up to visit those
teams. He's not the "God of Linux"- he just has final say as to what
goes into the kernel that calls itself "Linux". Everything outside the
kernel is fair game for anyone to slap together and call a
distribution.
Which would you rather have- a distribution backed by RedHat or SuSe-
organizations that grew up around Linux- or a distribution backed by
IBM, an organization that arguably understands enterprise computing?
Something that scales from the desktop to the datacenter, backed by
your own wits or the best support organization there is?
> Right now you need support for Oracle databases if you want to be a
> Linux vendor. Oracle would be dead dumb if they helped their largest
> competitor on DB software getting a monopoly on the Linux platform.
Nobody's said anything about a monopoly.
> IBM has been spending much money on Linux, but it is difficult to see
> that they have got that much revenue from that. Linux fans may like
> what IBM do, but the fans do not let that decide their purchase
> decisions.
You're not looking hard enough. I can think of a fairly large former
DEC customer that's now selling Linux on IBM hardware into their
customers for all the new systems they sell. The hardware's good, and
they're running Linux because their people can make it do what they
need. (They're tired of being caught out with proprietary OSes, you
see, unable to do anything but *request* changes from the mothership,
and/or having their OS yanked out from under them (Tru64))
Main, Kerry
> -----Original Message-----
> From: DeanW [mailto:dean.w...@gmail.com]
> Sent: March 14, 2005 9:59 AM
> To: Info...@Mvb.Saic.Com
> Subject: Re: HP on fast track to finding a CEO
>
[snip...]
> > IBM has been spending much money on Linux, but it is
> difficult to see
> > that they have got that much revenue from that. Linux fans may like
> > what IBM do, but the fans do not let that decide their purchase
> > decisions.
>
> You're not looking hard enough. I can think of a fairly large former
> DEC customer that's now selling Linux on IBM hardware into their
> customers for all the new systems they sell. The hardware's good, and
> they're running Linux because their people can make it do what they
> need. (They're tired of being caught out with proprietary OSes, you
> see, unable to do anything but *request* changes from the mothership,
> and/or having their OS yanked out from under them (Tru64))
>
So what happens when the Customer modifies the kernel and some ISV
application breaks or behaves differently?
Who does the Customer point his finger at when the internal guy who made
the changes is gone on vacation and some other issue comes up and then
Red Hat / Suse discovers that custom kernel or distribution supplied
driver changes have been made.
As someone else mentioned, both open source and closed have benefits and
challenges.
For some Customers, running production support with limited vendor
support for their applications after they make internal changes in OS or
application is an acceptable risk. For others, the "one throat to choke"
is very much the way they are moving.
Different strokes ..
Dan Foster
>
> So what happens when the Customer modifies the kernel and some ISV
> application breaks or behaves differently?
No different from if the Customer had done something equivalent with a
closed source software. Customer would likely be told to revert first
before support would proceed, as is customary practice. Some additional
options may exist but these would be vendor-specific.
> For some Customers, running production support with limited vendor
> support for their applications after they make internal changes in OS or
> application is an acceptable risk. For others, the "one throat to choke"
> is very much the way they are moving.
Allow me to point out something:
Customers who are disciplined enough to run production systems on
closed source platforms will not likely suddenly pull a Jekyll-and-Hyde
situation and behave differently with open source platforms.
We run both types of platforms, and apply the same operational practices
and standards to both. After all, why would we treat operational
requirements any different regardless of platform type?
Anyone who is really that undisciplined enough to be loading unsupported
code versions (or code modifications) and needs external support, will
reap what they sow. This is true regardless of open or closed source.
Open source platforms *do* have their healthy share of issues, but this
particularly dubious example of FUD isn't one of them -- that is, not
specific to open source.
Might I suggest use of better examples to buttress your position? ;)
-Dan
JF Mezei
When the vendor stops supporting a product, with open source, you have
the option to continue to develop it yourself and make interfaces to
newer software.
Consider the various applications that come with TCPIP Services. they
are closed source and we can't do a thing to bring them up to par, and
with VMS engineering rumoured to be downsizing, the odds of getting
serious updates are slim.
But if these were open sourced, we could still put or add ones to the
receiver for isntance and integrate logic for better spam handling etc etc.
Similarly, if it were open sourced, we would have access to the varoius
TCPIP$ routines in the shareable image which woudl allow much better
interfaces t the smtp symbiont and name resolver for instance.
This is where Linux has got a real edge over the rest of the world.
NOW, if VMS were a strategic product and given the proper number of
resources, it COULD beat the pants off Linux with far better products.
But right now, the feeling I get is that VMS's tank is empty and it is
running on fumes with a skeleton engineering team (whose work is still
very impressive, and while they may ne our heros, they are still humans
and limited in what they do).
Main, Kerry
> -----Original Message-----
> From: Dan Foster [mailto:use...@evilphb.org]
> Sent: March 14, 2005 12:18 PM
> To: Info...@Mvb.Saic.Com
> Subject: Re: HP on fast track to finding a CEO
>
> <FD827B33AB0D9C4E92E...@tayexc19.americas.cpqc
> orp.net>, Main, Kerry <kerry...@hp.com> wrote:
> >
> > So what happens when the Customer modifies the kernel and some ISV
> > application breaks or behaves differently?
>
> No different from if the Customer had done something equivalent with a
> closed source software. Customer would likely be told to revert first
> before support would proceed, as is customary practice. Some
> additional
> options may exist but these would be vendor-specific.
Sigh... I likely will regret going down this murky road, but -
How many Customers can afford to roll back production environents to
reproduce problems that are only seen under heavy load etc? When Prod
problems happen, many Cust's need fixes with minimal impact on their
prod environment and "rolling" back is not an option.
It sounds like your environment is such that you are comfortable doing
your own support and that is fine. Great.
However, other Customers do not want to maintain OS and kernel driver
changes - they want to focus on other areas of the business and let the
vendors worry about these low level issues.
>
> > For some Customers, running production support with limited vendor
> > support for their applications after they make internal
> changes in OS or
> > application is an acceptable risk. For others, the "one
> throat to choke"
> > is very much the way they are moving.
>
> Allow me to point out something:
>
> Customers who are disciplined enough to run production systems on
> closed source platforms will not likely suddenly pull a
> Jekyll-and-Hyde
> situation and behave differently with open source platforms.
>
> We run both types of platforms, and apply the same
> operational practices
> and standards to both. After all, why would we treat operational
> requirements any different regardless of platform type?
>
We all know that no matter how much testing you do, many issues can
still occur in production environment. If you are comfortable resolving
these on your own or alternatively "rolling back", then that is
certainly one approach.
As much as possible, other Cust's want the vendors to fix these OS /
driver level issues.
> Anyone who is really that undisciplined enough to be loading
> unsupported
> code versions (or code modifications) and needs external support, will
> reap what they sow. This is true regardless of open or closed source.
>
> Open source platforms *do* have their healthy share of
> issues, but this
> particularly dubious example of FUD isn't one of them -- that is, not
> specific to open source.
>
> Might I suggest use of better examples to buttress your position? ;)
>
> -Dan
Actually, what we are really talking about is how much risk is the
Customer willing to take on them selves. Yes, there are some benefits of
Open Source, but there is also a cost in terms of technical resources
required to do this support (coding, testing, QA, documentation) and
potential downtime - especially in 24x7 environments.
Anyway, as I stated before, there are pro's-n-con's with both so called
closed vs open approaches and this is one debate that could (and likely
will) go one for decades.
b...@instantwhip.com
Main, Kerry wrote:
>
> Actually, what we are really talking about is how much risk is the
> Customer willing to take on them selves. Yes, there are some benefits
of
> Open Source, but there is also a cost in terms of technical resources
> required to do this support (coding, testing, QA, documentation) and
> potential downtime - especially in 24x7 environments.
what cost? If HP would keep a couple of "c" programmers to work
to exclusively port key opensource pieces to vms (i.e. apache)
that cost would be minimal ... and just think of the revenue
stream those 2 c programmers could generate!
Simon Clubley
>
> There are distributions out there, which differ from RedHat only in the
> logos.
The main one been Fedora, which is currently at version Fedora Core 3.
This is RedHat minus the logos, is distributed by RedHat, and RedHat
engineers participate in the RedHat run support forums (which I discovered
while having a problem installing it.)
I'm disappointed by Kerry's somewhat flippant arguments. For example,
people are unlikely to suddenly start doing inhouse kernel modifications
on Linux unless they did the same thing on VMS, and if they do, then the
support position would be the same as it would be on VMS.
Also, a convention in open source software, when you make changes that for
some reason or other are not integrated into the base code at once, is to
distribute those changes as a patch from the base code. If you want the
additional facilities, you use the patch. If you don't, then you don't use
the patch.
Over time, one of two things will happen: either the patch will be integrated
into the mainline or the enhancements will continue life as a separate patch.
For example, RTLinux (the real time extensions to Linux) are distributed
as a patch against the Linux kernel.
Another example, RTEMS, a open source RTOS from OAR, requires changes to
the compiler tools (both GCC and binutils) in order to build. Initially
these changes were distributed as full patches to the tools and over time
a number of them have been integrated into the mainline code.
Simon.
--
Simon Clubley, clubley@remove_me.eisner.decus.org-Earth.UFP
Microsoft: The Standard Oil Company of the 21st century
Main, Kerry
> -----Original Message-----
> From: b...@instantwhip.com [mailto:b...@instantwhip.com]
> Sent: March 14, 2005 1:29 PM
> To: Info...@Mvb.Saic.Com
> Subject: Re: HP on fast track to finding a CEO
>
>
Actually, I was referring to Customers maintaining their own technical
support staff for maintaining things like OS/driver/misc code.
HP does have a number (no, I do not know how many) of programmers
working on Open Source code for OpenVMS. As you mentioned, the Apache /
OpenVMS prod is one example. Other examples include security, browser
and source code products like Netbeans etc. Others as well.
There are also other resources like Turin University that are also doing
work in this area.
Reference:
http://h71000.www7.hp.com/openvms/integrity/integrity_video_script.html
(scroll down to Turin University comments)
"We have successfully ported security application to OpenVMS on
Integrity servers and realized a 30% performance gain in cryptographic
algorithms. In the next year the Politecnico will increase greatly its
size and scope. So also the computing infrastructure must be extremely
flexible, scalability, cost effective and preserving today's investment.
We are confident that HP can give us in the future as in the past all
these values."
Main, Kerry
> -----Original Message-----
> From: Simon Clubley
> [mailto:clubley@remove_me.eisner.decus.org-Earth.UFP]
> Sent: March 14, 2005 1:36 PM
> To: Info...@Mvb.Saic.Com
> Subject: [OT] Linux, was: Re: HP on fast track to finding a CEO
>
> In article <42354e3d$0$78284$157c...@dreader1.cybercity.dk>,
> Karsten Nyblad <nos...@nospam.nospam> writes:
> >
> > There are distributions out there, which differ from RedHat
> only in the
> > logos.
>
> The main one been Fedora, which is currently at version Fedora Core 3.
> This is RedHat minus the logos, is distributed by RedHat, and RedHat
> engineers participate in the RedHat run support forums (which
> I discovered
> while having a problem installing it.)
>
> I'm disappointed by Kerry's somewhat flippant arguments. For example,
> people are unlikely to suddenly start doing inhouse kernel
> modifications
> on Linux unless they did the same thing on VMS, and if they
> do, then the
> support position would be the same as it would be on VMS.
>
Simon,
They were not flippant remarks.
Previous replies stated things that stated one of the advantages of Open
Source was that you could do kernel / driver/ priv'ed level changes and
not wait on the vendor to make them. Certainly, there are advantages of
this capability, but I simply pointed out that there are issues with
this that need to be considered in terms of the level of risk the
company is willing to take on.
As I have been careful to say, there are advantages and disadvantages to
both approaches (open vs so called "closed") and one needs to consider
these when deciding which is best for your own environment.
Note - why is when anyone points out a disadvantage of open source, that
almost immediately all of the open source promoters come rushing in and
defending Open Source as if that one specific criticism was somehow an
attack on Open Source overall?
John Smith
And quality of people too. There are precious few people I'd let muck around
with a kernel vs. the number who'd be ok at application level coding or
report writing.
--
OpenVMS - The never advertised operating system with
the dwindling ISV base.
John Smith
Come on Kerry.....Turin's old news....how about some others participating in
the programme.
Come to thing of it, how about the latest $50MM OpenVMS deals too. After
all...VMS is growing by leaps and bounds (double digits....or did that
really just mean that HP sold somewhere between 10-99 VMS servers last
year?)
David J Dachtera
> [snip]
> [quoting from an article]
> Asked whether IBM's eventual goal is to replace AIX with Linux, Mills
> responded, "It's fairly obvious we're fine with that idea...It's the
> logical successor."
Hhmmm... Wonder what VMS's "logical successor" would be...
> "OpenVMS has always had integrity ..
Well, perhaps, yes - VMS has had more of it than its owners.
> Now, Integrity has OpenVMS .."
Hhmmm... I wonder if "Integrity Systems, Inc." is registered
(registerable?) as a trade mark...?
--
David J Dachtera
dba DJE Systems
http://www.djesys.com/
Unofficial OpenVMS Hobbyist Support Page:
http://www.djesys.com/vms/support/
Unofficial Affordable OpenVMS Home Page:
http://www.djesys.com/vms/soho/
Unofficial OpenVMS-IA32 Home Page:
http://www.djesys.com/vms/ia32/
Coming soon:
Unofficial OpenVMS Marketing Home Page