British Survey Shows that Linux is Gaining Ground on Windows
Roy Schestowitz
|
| The days are over when Microsoft could rely on being the only
| software platform in town. John Dwyer reports
|
| Microsoft has a case to answer. When Quocirca sifted through 8000
| IT professionals' responses to an online questionnaire last year, the
| UK consultancy discovered deep unease with Microsoft?s Windows operating
| system.
|
| [...]
`----
Do read on:
http://www.themanufacturer.com/britishindustry/content_page.html?article_id=559
Rex Ballard
Larry Qualig
Good article. Interesting how it ends:
- "If it's of any interest, after years in Windows wilderness John
Dwyer (the author) now prefers the serenity of the Unix-world Apple Mac
to the excitements of Mr Gates's product."
John Bailo
> http://www.themanufacturer.com/britishindustry/content_page.html?article_id=559
>
This article is interesting, but I think there should be a rule about
saying things like "Gaining Ground" when it refers only to opinion and
not actual numbers of installed systems (goes for both sides).
Roy Schestowitz
"Ground" does not have to be concrete in a loose context such as this. See
Larry's mind-provoking discussion on counting Linux users. I am more than
convinced that Linux is gaining in substaintial (quantitatively) ground; and
not only in the server room...
John Bailo
> I am more than
> convinced that Linux is gaining in substaintial (quantitatively) ground; and
> not only in the server room...
Me, too.
And, in some sense, Linux is not only a criticism of the previous
generation of software, but also a criticism of the institutions that
wrote that software, the way it was marketed and the communications
channels through which it was promoted and reported.
So, one has to take the whole thing into question and consider what
really matters about Linux, per se.
billwg
"Roy Schestowitz" <newsg...@schestowitz.com> wrote in message
news:dv9du5$2a5s$2...@godfrey.mcc.ac.uk...
I particularly liked the part that said:
"Blowers agrees: "Too much has been made of the cost of Microsoft
licences. After all, in return for that you get support. With Linux you
can't afford to go without support and you'll have to get it from third
parties like Novell or Red Hat." "
Ray Ingles
> I particularly liked the part that said:
>
> "Blowers agrees: "Too much has been made of the cost of Microsoft
> licences. After all, in return for that you get support. With Linux you
> can't afford to go without support and you'll have to get it from third
> parties like Novell or Red Hat." "
I'm sure you did. My favorite part:
But Tarpey thinks the support problem is "overstated": "There are
risks," he agrees. "You have to be careful where you get your code from.
Some of it is ropey. But with Microsoft you pay for the service." Every
call you make to Microsoft has to be paid for and, if you’re a small,
cash-strapped Windows user, "you're in the same boat."
A few moments' browsing reveals plentiful free support from
Linux-community websites. In the Linux world, someone somewhere has
already encountered your problem and published a solution.
As Latham acknowledges, no-one should expect to run Windows without
in-house expertise either, "because of the security issue." Every
Windows-based business, he says, needs someone to make sure the software
updates and anti-virus software are installed and up to date: "To keep
yourself secure you need an IT resource."
--
Sincerely,
Ray Ingles (313) 227-2317
"Engineering is like having an 8 a.m. class and a late afternoon
lab every day for the rest of your life." - Anonymous
Mr X
> to earth, dirt or soil that you walk on. So literally it
> could mean "...". Possibly an off-handed
> reference to "...having dirt thrown in ones face".
>
> I think it meant Linux eats more dirt.
>
>
You are truly an anathema aNA tHEMA
With your truly insightful analysis of Linux.. maybe I should go back to
that crappy MsOFT Siht.......
On Second thoughts.......
No, your argument doesn't hold enough ground. Tell me why should I
abandon this beautifull Linux Envioroment that runs my business so
wonderfully, so hassle free, I've even moved my Clipper Dbase with the
help of www.xharbour.org so my POS software runs without a hitch.
aNA tHEMA
or whatever you prefer to call yourself
getting more dirt
having dirt thrown in ones face
refers to you.
Roy Schestowitz
> On 2006-03-15, billwg <bi...@twcf.rr.com> wrote:
>> I particularly liked the part that said:
>>
>> "Blowers agrees: "Too much has been made of the cost of Microsoft
>> licences. After all, in return for that you get support. With Linux you
>> can't afford to go without support and you'll have to get it from third
>> parties like Novell or Red Hat." "
>
> I'm sure you did. My favorite part:
Why is he being fed?
> But Tarpey thinks the support problem is "overstated": "There are
> risks," he agrees. "You have to be careful where you get your code from.
> Some of it is ropey. But with Microsoft you pay for the service." Every
> call you make to Microsoft has to be paid for and, if you're a small,
> cash-strapped Windows user, "you're in the same boat."
>
> A few moments' browsing reveals plentiful free support from
> Linux-community websites. In the Linux world, someone somewhere has
> already encountered your problem and published a solution.
> As Latham acknowledges, no-one should expect to run Windows without
> in-house expertise either, "because of the security issue." Every
> Windows-based business, he says, needs someone to make sure the software
> updates and anti-virus software are installed and up to date: "To keep
> yourself secure you need an IT resource."
That's where the 'learning curve' comes into play. While I am highly skilled
with all versions of Windows (down to the level of keyboard accelerators), I
have always refused to learn how to 'protect' Windows. When colleagues brag
about their 'knowledge' on Windows security, I make them aware of the fact
that it is no IT skill. It is the skill of making up for design bugs of one
particular O/S. Sadly, it's a necessary survival skill nowadays, for Windows
users only.
Sinister Midget
> I'd love to read it, but it looks like the link is gone.
Works here. 0403 CST.
--
Bobax: Innovative Microsoft peer-to-peer software.
Sinister Midget
But the support comes at a price. Latham says some K3 sites moved to
Linux when Microsoft increased its licence fees and introduced
subscription charges. Some companies suddenly faced a huge increase
? £10,000 to £15,000 a year ? that they hadn?t bargained for just
to continue using their SQL databases, Exchange email software and
other Microsoft code. And if they consulted a Microsoft help line,
that cost extra.
You pay for software, you get garbage. You pay for support, you have to
pay more to really get any support(1).
No wonder Blammer isn't ashamed to act idiotic in public. As long as
His Buttcrustness spews out meaningless catch-phrases and the droolers
lap it up, those criminals can skin the "customers" (aka "marks")
without even trying to hide what they're doing.
(1) And that's most likely to be "Jerome" in Aurangabad, shooting the
breeze with you like a long-lost buddy, while you wait, and wait, and
wait for the umpteenth reboot to finish, knowing full well that several
reboots from now will leave you, at best, in the same condition you
were in before this all started.
--
How dare the government intervene to stifle innovation in the computer
industry! That's Microsoft's job, dammit!
billwg
"Ray Ingles" <sorc...@localhost.localdomain> wrote in message
news:slrne1h269....@localhost.localdomain...
>
> I'm sure you did. My favorite part:
>
> But Tarpey thinks the support problem is "overstated": "There are
> risks," he agrees. "You have to be careful where you get your code
> from.
> Some of it is ropey. But with Microsoft you pay for the service."
> Every
> cash-strapped Windows user, "you're in the same boat."
>
"One obvious cost case for Linux is among unix-based organizations who
want to reduce hardware costs. Butler Group analyst Mark Blowers says
Linux allows them to move from proprietary servers from IBM, HP or Sun
to cheaper, standard Intel boxes."
But the most incisive was:
"Andy Latham, technical director of SME applications supplier K3's IEG
offshoot, confirms this: "Our customers are not talking to us about
Linux," he says. "They want a solution to business problems."
Latham's users start from Microsoft Office for the desktop and then ask
what email and other business applications, even CAD, work with that.
From their position, says Latham, "the Microsoft solution is the only
one. and it all works in a Windows environment." "
Along with:
"Mike Evans, senior partner at industrial IT consultancy Cambashi, says
"a lot of rubbish is talked about security." Most security failures have
little to do with technology and everything to do with how people
behave: "I don't think it's a significant differentiator. It's a
licensing, sourcing argument."
Now you can read the tea leaves anyway you want, ray, but from my POV,
linux is sapping the strength from unix and that can only be a plus for
Mr. Softee. People who like to do sophisticated applications like CAD
are able to make some money at the game, too, so why do it for free or
cheap? At the end of the day, the hundred or so bucks for workstation
Windows is nothing compared to the thousands per seat for the
applications running on it and who's going to want to trust that to
chance?
Ray Ingles
> Now you can read the tea leaves anyway you want, ray, but from my POV,
> linux is sapping the strength from unix and that can only be a plus for
> Mr. Softee.
I don't see it as "sapping the strength from unix" at all. I see it as
making Unix vastly more cost-effective. Even the article backs that up.
> At the end of the day, the hundred or so bucks for workstation
> Windows is nothing compared to the thousands per seat for the
> applications running on it and who's going to want to trust that to
> chance?
What "chance" are you talking about?
--
Sincerely,
Ray Ingles (313) 227-2317
"The roles of giant, transnational corporations and government have
slowly reversed. Government is now more an instrument of such
corporations than the corporations are instruments of government."
- Dee Hock, founder of VISA
billwg
"Ray Ingles" <sorc...@localhost.localdomain> wrote in message
> On 2006-03-16, billwg <bi...@twcf.rr.com> wrote:
>> Now you can read the tea leaves anyway you want, ray, but from my
>> POV,
>> linux is sapping the strength from unix and that can only be a plus
>> for
>> Mr. Softee.
>
> I don't see it as "sapping the strength from unix" at all. I see it as
> making Unix vastly more cost-effective. Even the article backs that
> up.
>
think that it is cool for a bunch of developers to develop for the love
of mankind and applaud what they have done, but you are out to lunch.
There have been a lot of claims lately about linux starting in 1991
which I think is premature, but in any case, Windows, starting around
the same time, has made a giant dent in the traditional software
platform businesses whereas linux is only starting and that solely
because of the effort being put into it by IBM.
The problem with linux is that it needs the continued volunteerism
effort to even marginally compete and there is no guarantee that the
boys will stay down on the farm.
>> At the end of the day, the hundred or so bucks for workstation
>> Windows is nothing compared to the thousands per seat for the
>> applications running on it and who's going to want to trust that to
>> chance?
>
> What "chance" are you talking about?
>
The chance that the platform developers manage to stay with the needs of
the application developers. Look what happened to Unix and VAX
workstation apps.
Rex Ballard
The Linux based revenues of Red Hat and Novell and Mandriva and
Linspire are based almost entirely on support contracts. These vendors
compete with each other, which means that they have to provide much
better support at very competitive prices.
Microsoft offers a "support package" which seems to consist mainly of
the online update service included in with the OEM version of the
operating system, and then offers corporate customers "support
contracts" in which the company is responsible for level 1/2 support
which includes restarting the app, rebooting the box, reinstalling the
application, reinstalling Windows, reformatting the hard drive and
reinstalling the "Standard Load". By then, who needs Microsoft?
Most Linux distributors try to support not only the operating system
but a number of applications including OSS applications, third party
ISV applications, and custome applications designed to make system and
software management easier.
In the server market, Linux has been displacing Microsoft (nearly 45%
of all new Linux deployments are replacements of Microsoft systems)
precisely because the cost of maintainance, backups, support, recovery,
management - are all so much lower with Linux. Linux is based on
standards and technologies which were designed to be supported using
remote access, scripting tools and automated "cron" jobs, with minimal
manual ineraction and system "reboots" only necessary when completely
replacing the kernel for a major upgrade.
There are many Linux systems that get stuck in a cool closet, a back
room, or an isolated rack - and don't even have a keyboard plugged in.
Such systems can go for years without direct physical contact and
without requiring significant console interactions.
Microsoft has tried to step up to the plate with "Terminal Server" and
"Remote Access" but this is still a GUI console which can be very labor
intensive. In addition, most Windows servers are only licensed for 2
concurrent user accesses - which is almost never enough during a
problem resolution exercise or a system configuration excercise..
The only system with longer uptimes than Linux is FreeBSD and this is
primarily because there has been less interest in things like
supporting SMP, NUMA, and dual-core processors and less interest in
supporting 64 bit processors.
Many companies report "across the board" reductions in TCO of as much
as 85% when switching from Windows to Linux.
Microsoft likes to cite a handful of companies who had no Linux
expertise on site, made no attempt to contract with 3rd party support
services (Many ISPs offer Linux support to SMBs), and sent people to
formal "cram courses" at full pay and full price, usually off-site, and
added those costs in, while making no allowance for similar training
costs, such as recertification of MCSEs, for Windows. Furthermore, at
least one of these companies sent their users to classes that included
topics like the "vi" editor, "bash" shell programming, and PERL
programming. Classes intended for Web Site administrators, not end
users of desktop Linux.
Still, the article was interesting - though it did look like it had
been through the "Microsoft Revision Process".
chrisv
>The problem with linux is that it needs the continued volunteerism
>effort to even marginally compete and there is no guarantee that the
>boys will stay down on the farm.
There $billions in corporate money being invested in Linux, and
there's zero reason to believe that the volunteers will dry-up either.
Plus, with China and other countries committing to open source...
Look out Billy.
Ray Ingles
>> I don't see it as "sapping the strength from unix" at all. I see it as
>> making Unix vastly more cost-effective. Even the article backs that
>> up.
>>
> Well, you have no regard for the effectiveness of financing, ray. You
> think that it is cool for a bunch of developers to develop for the love
> of mankind and applaud what they have done, but you are out to lunch.
Gee, what are these words doing in my mouth? Who put them there? Oh,
billwg. No wonder they taste so bad.
People make money on the Internet despite the technology being open
and, indeed, mostly open-source. The same can be - and is - true for
Linux. Why do you assume that the "financing" will all just vanish in a
puff of smoke? When you don't have to pay monopolistic prices for
infrastructure, the money can be better spent on other development.
> There have been a lot of claims lately about linux starting in 1991
> which I think is premature
That's funny. It's a matter of public record and all, but you can call
it "premature" if you wish. The amusing thing is, that's really pretty
late, as the infrastructure like the compilers and apps had already been
developed. (See Stallman et. al.)
> but in any case, Windows, starting around
> the same time, has made a giant dent in the traditional software
> platform businesses whereas linux is only starting and that solely
> because of the effort being put into it by IBM.
It's been growing rapidly - exponentially, really - since 1998 and is
now around 10% of server revenue, despite dramatically lower prices.
And IBM isn't getting all of that money...
> The problem with linux is that it needs the continued volunteerism
> effort to even marginally compete
Wait, I thought it was IBM's commercial contributions? Geez, try to
keep your stories straight, at least within the same post!
> and there is no guarantee that the boys will stay down on the farm.
Aside from the fact that your premise is wrong - plenty of technologies
can be and are developed in an open manner that combines both commercial
and "volunteer" activities to the benefit of both (e.g. TCP/IP, OpenGL,
HTTP/HTML, etc.) - can you point to any indication whatsoever that
there's any risk of "the boys" *not* staying "down on the farm"?
>>> At the end of the day, the hundred or so bucks for workstation
>>> Windows is nothing compared to the thousands per seat for the
>>> applications running on it and who's going to want to trust that to
>>> chance?
>>
>> What "chance" are you talking about?
>>
> The chance that the platform developers manage to stay with the needs of
> the application developers. Look what happened to Unix and VAX
> workstation apps.
Why don't you enlighten me on exactly what you think happened?
--
Sincerely,
Ray Ingles (313) 227-2317
"Yes, the long war on Christianity. I pray that one day we may live in an
America where Christians can worship freely, in broad daylight, openly
wearing symbols of their religion, perhaps around their necks. And maybe -
dare I dream it - maybe one day there could even be an openly Christian
president. Or, perhaps, 43 of them. Consecutively." - Jon Stewart
hanum...@gmail.com
"And [Linux is] used by a third of UK local authorities, though
that's still way behind France's 71, Germany's 68 and Holland's
55 per cent."
Really? Christ almighty!
billwg
"Ray Ingles" <sorc...@localhost.localdomain> wrote in message
> On 2006-03-16, billwg <bi...@twcf.rr.com> wrote:
>>> I don't see it as "sapping the strength from unix" at all. I see it
>>> as
>>> making Unix vastly more cost-effective. Even the article backs that
>>> up.
>>>
>> Well, you have no regard for the effectiveness of financing, ray.
>> You
>> think that it is cool for a bunch of developers to develop for the
>> love
>> of mankind and applaud what they have done, but you are out to lunch.
>
> Gee, what are these words doing in my mouth? Who put them there? Oh,
> billwg. No wonder they taste so bad.
>
noble cause, beyond the pale of mere profit. You can't back down from
having said that, as silly as it was to do so.
> People make money on the Internet despite the technology being open
> and, indeed, mostly open-source. The same can be - and is - true for
> Linux. Why do you assume that the "financing" will all just vanish in
> a
> puff of smoke? When you don't have to pay monopolistic prices for
> infrastructure, the money can be better spent on other development.
>
$50 is "monopolistic"? Tut, ray, its an artificially low price to cut
off Netscape's air supply on one hand and monopolistic on the other with
you guys.
>> There have been a lot of claims lately about linux starting in 1991
>> which I think is premature
>
> That's funny. It's a matter of public record and all, but you can call
> it "premature" if you wish. The amusing thing is, that's really pretty
> late, as the infrastructure like the compilers and apps had already
> been
> developed. (See Stallman et. al.)
>
Well, OK. So that means that linux is a small fraction as effective as
Windows, having had the same time to sway users. The OSS hand to mouth
approach has produced a weak effort, resulting in some business level
about one third of Microsoft's in the small server market and about a
hundredth of Microsoft's in the vast desktop market. As a pure play it
is barely a measureable fraction of one percent of Microsoft's market
share. I'd rate that as a loser personally.
>> but in any case, Windows, starting around
>> the same time, has made a giant dent in the traditional software
>> platform businesses whereas linux is only starting and that solely
>> because of the effort being put into it by IBM.
>
> It's been growing rapidly - exponentially, really - since 1998 and is
> now around 10% of server revenue, despite dramatically lower prices.
> And IBM isn't getting all of that money...
>
Too little, too late, ray.
>> The problem with linux is that it needs the continued volunteerism
>> effort to even marginally compete
>
> Wait, I thought it was IBM's commercial contributions? Geez, try to
> keep your stories straight, at least within the same post!
>
Not a problem, ray. IBM glommed onto linux as a mechanism to facilitate
a strong pricing move against Sun Microsystems. Nothin else. You can't
cut prices a lot and not show your old customers a rational reason why
or else they will figure that you have been screwing them all along.
That's a fundamental premise of commercial sales. So here was the
excuse, i.e. the IBM servers with linux could do most of the unix jobs
that IBM was competing for and save the cost of AIX and so allow for the
lower price on the same hardware. That kept the price for the AIX
system where it was and got the camel's nose in the unix tent.
But it couldn't happen if linux had to pay for a lot of development
staff.
>> and there is no guarantee that the boys will stay down on the farm.
>
> Aside from the fact that your premise is wrong - plenty of
> technologies
> can be and are developed in an open manner that combines both
> commercial
> and "volunteer" activities to the benefit of both (e.g. TCP/IP,
> OpenGL,
> HTTP/HTML, etc.) - can you point to any indication whatsoever that
> there's any risk of "the boys" *not* staying "down on the farm"?
>
What have they done for us lately, ray? All I see is the OSS bunch
milling around and trying to clone the commercial products or at least
add the popular characteristics as their vision of what is hot and what
is not gets clearer with time.
>>>> At the end of the day, the hundred or so bucks for workstation
>>>> Windows is nothing compared to the thousands per seat for the
>>>> applications running on it and who's going to want to trust that to
>>>> chance?
>>>
>>> What "chance" are you talking about?
>>>
>> The chance that the platform developers manage to stay with the needs
>> of
>> the application developers. Look what happened to Unix and VAX
>> workstation apps.
>
> Why don't you enlighten me on exactly what you think happened?
>
Ask Intergraph and ComputerVision how a reliance on UNIX's "superior OS"
worked out in the long run!
Rex Ballard
billwg wrote:
> "Ray Ingles" <sorc...@localhost.localdomain> wrote in message
> news:slrne1h269....@localhost.localdomain...
> >
> > I'm sure you did. My favorite part:
> >
> > But Tarpey thinks the support problem is "overstated": "There are
> > risks," he agrees. "You have to be careful where you get your code
> > from. Some of it is ropey. But with Microsoft you pay for the service."
> > small, cash-strapped Windows user, "you're in the same boat."
You have to be careful with when you get code for Windows too. At
least if I get 5,000 packages from SUSE on a DVD, I have a pretty good
idea that they have been at least validated to be functional and free
of trojans.
If I downloaded 5,000 shareware programs for Windows, my chances of
getting at least a few viruses are very high. In fact, I probably only
have a 1 in 500 change of NOT getting something nasty.
> Of course there was the "obvious":
>
> "One obvious cost case for Linux is among unix-based organizations who
> want to reduce hardware costs. Butler Group analyst Mark Blowers says
> Linux allows them to move from proprietary servers from IBM, HP or Sun
> to cheaper, standard Intel boxes."
Organizations that already have UNIX are definitely the best candidates
- which pretty much means nearly all of the companies with over 200
employees or over $100 million in annual revenue. And as much as they
might consider migrating UNIX machines to Linux, they are even MORE
eager to migrate WINDOWS machines to Linux, especially Windows NT 4.0
machines - which they were licensed for about $1500 per CPU, and are
now being told to upgrade to Windows 2003 - at a cost of over $25,000
per CPU.
During the height of the push to Windows NT 4.0 migration, many
companies had as many as 300 windows machines per thousand employees.
Most of these were 4 processor machines.
Many organizations have switched to fewer big Windows machines and have
been migrating as much as possible from NT 4.0 to Linux. In addition,
most new applications and servers are now being requested on Linux as
the customers FIRST choice. The second choice is traditional UNIX such
as Solaris, AIX, and HP_UX.
The UNIX box counts are going down because the average UNIX box of
today typically has 64 dual-core processors with bandwidths in hundreds
of gigabytes per second compared to machines of just 10 years ago that
were usually 4 processors with bandwidths of only a few hundred
megabytes per second. In addition, logical partitioning,
micropartitioning, and other load management techniques have made it
easier to put lots more "servers" on a single physical box. A top of
the line StarFire is the equivalent of almost 1000 Sparc20 servers.
It's the equivelant of about 4000 NT 4.0 single-processor servers from
1997 (when NT 4.0 was first being adopted)..
> But the most incisive was:
>
> "Andy Latham, technical director of SME applications supplier K3's IEG
> offshoot, confirms this: "Our customers are not talking to us about
> Linux," he says. "They want a solution to business problems."
> Latham's users start from Microsoft Office for the desktop and then ask
> what email and other business applications, even CAD, work with that.
> From their position, says Latham, "the Microsoft solution is the only
> one. and it all works in a Windows environment." "
Consider the source. Yes it's true - the pointed headed bosses
(Dilbert) just want Windows, some great video games, and maybe a good
stock portfolio tracker. The CAD guys want UNIX or Linux on their
desktops - because the high-end CAD systems are all based on Unix.
Most of the high-end CAD packages have also been ported to Linux.
Most of those companies who are implementing Linux on desktop are doing
so because their employees are demanding it, and/or because their
competitors are using it and beating the feathers of them. Let's face
it, if Linux only saves you 30% of the cost of Windows, that's like
having about 1/3 of your profit to reinvest in strategic initiatives
which will increase sales and lower costs - increasing profits.
How much do you REALLY stand to gain by upgrading to VISTA? or Office
2003? or whatever ELSE Microsoft wants to shove down your CIO's throat.
> Along with:
>
> "Mike Evans, senior partner at industrial IT consultancy Cambashi, says
> "a lot of rubbish is talked about security." Most security failures have
> little to do with technology and everything to do with how people
> behave: "I don't think it's a significant differentiator. It's a
> licensing, sourcing argument."
Great, you've just told 800 million people who have purchased Windows
machines, run Windows, and get infected by viruses 2-3 times/year that
they are getting viruses because THEY are IDIOTS! I wish I could get
an official statement from Bill Gates - on the record - to that effect.
People might just switch to Linux out of spite. Oops - they already
are.
> Now you can read the tea leaves anyway you want, ray, but from my POV,
> linux is sapping the strength from unix and that can only be a plus for
> Mr. Softee.
The problem is that many Linux projects become UNIX projects. I can
start with a small project that runs on Linux, Apache, MySQL, and PHP,
but as the project scales up, you might want to use Oracle on AIX or
Solaris instead of MySQL on Linux. When content management and
back-end integration get critical you might want to upgrade from PHP on
Linux to WebSphere on Solaris, which happens to come with IBM's version
of apache (the open source version - with a cute GUI for configuraing
the control files).
If I start with a WINDOWS server, the transition to UNIX is traumatic
at best. Failure to plan for that scalability of transitioning to
Linux or UNIX can be even more traumatic. Some Windows projects are
still-born because they can't even safely scale to full production.
Even Microsoft depends on a bank of Linux servers to provide load
balancing and overload protection as well as security for the Microsoft
websites, including MSN, CNBC, MSNBC, and Microsoft.com. This Linux
front-end means that the back-ends (hundreds of them) can crash to
their hearts content as box-booters run from crash to crash rebooting
the boxes.
> People who like to do sophisticated applications like CAD
> are able to make some money at the game, too, so why do it for free or
> cheap? At the end of the day, the hundred or so bucks for workstation
> Windows is nothing compared to the thousands per seat for the
> applications running on it and who's going to want to trust that to
> chance?
And that's just the point. If you're going to do a CAD package, you
can do like TurboCAD and make pretty pictures which can't feed the
robots and manufacturing equipment, or AutoCAD which create content
that can be fed into the REAL CAD systems that run on *NIX, the ones
that connect directly into the manufacturing equipment, with only a few
hundred errors/drawing.
Larry Qualig
Rex Ballard wrote:
> billwg wrote:
> > "Ray Ingles" <sorc...@localhost.localdomain> wrote in message
> > news:slrne1h269....@localhost.localdomain...
> > >
>
> companies had as many as 300 windows machines per thousand employees.
> Most of these were 4 processor machines.
>
> Many organizations have switched to fewer big Windows machines and have
> been migrating as much as possible from NT 4.0 to Linux. In addition,
> most new applications and servers are now being requested on Linux as
> the customers FIRST choice. The second choice is traditional UNIX such
> as Solaris, AIX, and HP_UX.
>
> The UNIX box counts are going down because the average UNIX box of
> today typically has 64 dual-core processors with bandwidths in hundreds
> of gigabytes per second compared to machines of just 10 years ago that
> were usually 4 processors with bandwidths of only a few hundred
> megabytes per second. In addition, logical partitioning,
> micropartitioning, and other load management techniques have made it
> easier to put lots more "servers" on a single physical box. A top of
> the line StarFire is the equivalent of almost 1000 Sparc20 servers.
> It's the equivelant of about 4000 NT 4.0 single-processor servers from
> 1997 (when NT 4.0 was first being adopted)..
[ snip ]
--> the average UNIX box of today typically has 64 dual-core
processors...
The "average" Unix box has 64 processors (dual-core)? No way, no how.
A Unix machine with 64-processors is anything but average.
Rex Ballard
isn't that unusual.
Most Z-Series machines are in the 32-64 core range - Linux on Z-Series
- lots of little VMs on a box. More common though are the Linux VM
shared with MVS, CICS, and the typcial mainframe VMs
Roy Schestowitz
Yes, I noticed that too. I wonder how Finland is doing. They are always far,
far ahead in that respect.
Best wishes,
Roy
--
Roy S. Schestowitz | "Disk quota exceeded; sig discontinued"
http://Schestowitz.com | SuSE Linux Ś PGP-Key: 0x74572E8E
11:30am up 9 days 4:07, 7 users, load average: 0.92, 0.82, 0.60
http://iuron.com - help build a non-profit search engine
chrisv
>"Ray Ingles" <sorc...@localhost.localdomain> wrote:
>>
>> Gee, what are these words doing in my mouth? Who put them there? Oh,
>> billwg. No wonder they taste so bad.
>>
>Oh, pshaw, ray. You are ever on about how giving to the community is a
>noble cause, beyond the pale of mere profit.
Idiot. That's not the same as what you claimed.
>You can't back down from
>having said that, as silly as it was to do so.
How ironic, considering the "silly" (read: incredibly fsching stupid)
things you are always saying.
>> People make money on the Internet despite the technology being open
>> and, indeed, mostly open-source. The same can be - and is - true for
>> Linux. Why do you assume that the "financing" will all just vanish in
>> a puff of smoke? When you don't have to pay monopolistic prices for
>> infrastructure, the money can be better spent on other development.
>>
>$50 is "monopolistic"? Tut, ray, its an artificially low price to cut
>off Netscape's air supply on one hand and monopolistic on the other with
>you guys.
Dumbsh*t. A typical office PC has a lot more than $50 of crapware on
it.
>> That's funny. It's a matter of public record and all, but you can call
>> it "premature" if you wish. The amusing thing is, that's really pretty
>> late, as the infrastructure like the compilers and apps had already
>> been
>> developed. (See Stallman et. al.)
>>
>Well, OK. So that means that linux is a small fraction as effective as
>Windows, having had the same time to sway users. The OSS hand to mouth
>approach has produced a weak effort,
If Linux is a "weak effort", what is the security nightmare Windwoes?
Total garbage, like you? LOL!!!
>resulting in some business level
>about one third of Microsoft's in the small server market and about a
>hundredth of Microsoft's in the vast desktop market.
You are lying again, billwg.
>As a pure play it
>is barely a measureable fraction of one percent of Microsoft's market
>share. I'd rate that as a loser personally.
Yet more obvious lies. I'd rate you as a loser personality. LOL!!!
Ray Ingles
>> Gee, what are these words doing in my mouth? Who put them there? Oh,
>> billwg. No wonder they taste so bad.
>>
> Oh, pshaw, ray. You are ever on about how giving to the community is a
> noble cause, beyond the pale of mere profit. You can't back down from
> having said that, as silly as it was to do so.
I've never said that. Go ahead, produce a cite of me saying that. I
triple-dog-dare you.
Giving to the community *is* a good thing, and enjoyable. But it's
*orthogonal* to a profit motive. As I've stated, numerous times, it's
entirely possible - indeed, trivial - to accomplish both at once. As I
have stated before, my own open-source contributions have (a) helped me
land jobs, and (b) directly provided me thousands of dollars from the
VA Linux IPO.
>> When you don't have to pay monopolistic prices for
>> infrastructure, the money can be better spent on other development.
>>
> $50 is "monopolistic"? Tut, ray, its an artificially low price to cut
> off Netscape's air supply on one hand and monopolistic on the other with
> you guys.
See this article:
http://www.it-director.com/article.php?articleid=13196
"In 1994 the TCP/IP stack and utility business for PCs was probably
about a $700 million affair. In 1995, it nearly evaporated when TCP/IP
begin to be natively supported in Windows 95. Thus the stack became
discreetly priceless when it was bundled into the ubiquitous OS. In many
ways, Open Source software is making the same demands on the
marketplace - these technologies are priceless, therefore stop trying to
make money them, but instead invest those same dollars in adding value
on top of the priceless technology. As a result, freely distributable,
standards based, basic technology will be a given, let's innovate on top
of it, where the real value, and may I add, margins, will be found."
The idea being that basic technologies like operating systems should be
a given, and innovation (even for - gasp! - profit) should happen on top
of that layer. Yes, I think this will happen. Just as TCP/IP made too
much sense for customers, vendors be damned, OSS like Linux *also* makes
too much sense for customers.
>> The amusing thing is, that's really pretty late, as the infrastructure
>> like the compilers and apps had already been developed. (See Stallman
>> et. al.)
>>
> Well, OK. So that means that linux is a small fraction as effective as
> Windows, having had the same time to sway users.
I'd say it's been *many times* as effective as Windows, seeing as it's
grown so well despite having a tiny fraction of the marketing budget of
Microsoft. Is still growing, in fact. Rapidly.
> The OSS hand to mouth approach has produced a weak effort, resulting
> in some business level about one third of Microsoft's in the small
> server market
["Small" being "$25K or less", and by *far* the fastest-growing segment
of the server market... but when billwg's counting, who cares?]
> and about a hundredth of Microsoft's in the vast desktop market.
It's at least equal to Apple's desktop share by anyone's count, and
that's with a marketing budget far less than Apple's.
> As a pure play it is barely a measureable fraction of one percent of
> Microsoft's market share. I'd rate that as a loser personally.
But Microsoft's share is stagnating - they can't really grow in their
given markets and they can't expand effectively into the developing
world, their overhead is far too high. Meanwhile, Linux is growing,
fast, not merely in traditional Microsoft markets but also in those
areas MS can't touch.
>> It's been growing rapidly - exponentially, really - since 1998 and is
>> now around 10% of server revenue, despite dramatically lower prices.
>> And IBM isn't getting all of that money...
>>
> Too little, too late, ray.
People usually say that about something that's *declining*, not growing
by double digit percentages every quarter. But we already knew you
inhabit a special world full of magical definitions of words.
>>> The problem with linux is that it needs the continued volunteerism
>>> effort to even marginally compete
>>
>> Wait, I thought it was IBM's commercial contributions? Geez, try to
>> keep your stories straight, at least within the same post!
>>
> Not a problem, ray. IBM glommed onto linux as a mechanism to facilitate
> a strong pricing move against Sun Microsystems. Nothin else.
Funny, nobody else sees it that way. From the same article I cited
above:
"Not too surprisingly, when one looks beyond the hype of the moment,
much of IBM's position on Open Source seems to be following the scenario
that I laid out. It is unlikely that Big Blue will suddenly cease to
have any software revenue, in fact I would argue that it will ultimately
rise as the company expands it reach and boosts the market overall
through its many open source contributions. While IBM is not the only
vendor that appears to have seen the logic in this course of action, it
is clearly the largest. With the Armonk Argonaut on board, it is likely
that much of the industry will follow, and will eventually not need to
talk about Open Source anymore than it does key foundational
technologies."
> But it couldn't happen if linux had to pay for a lot of development
> staff.
So, Linux's success requires *both* 'volunteer' *and* 'paid'
development?
Bravo, Bill! You've actually admitted something obvious! Linux succeeds
because it's neutral territory that *everyone* can exploit and benefit
from!
>> Aside from the fact that your premise is wrong - plenty of technologies
>> can be and are developed in an open manner that combines both commercial
>> and "volunteer" activities to the benefit of both (e.g. TCP/IP, OpenGL,
>> HTTP/HTML, etc.) - can you point to any indication whatsoever that
>> there's any risk of "the boys" *not* staying "down on the farm"?
>>
> What have they done for us lately, ray? All I see is the OSS bunch
> milling around and trying to clone the commercial products or at least
> add the popular characteristics as their vision of what is hot and what
> is not gets clearer with time.
Well, just in Linux itself:
http://www.kniggit.net/wwol26.html
Then there's the Xen virtualization technology, SELinux development,
futexes and the whole NPTL effort, etc. Again, that's just in the kernel
proper. See, e.g., the announcement of the new Gnome features for some
other areas of rapid improvement and innovation. Oh, and then there's
Firefox, which innovated nearly all of the new features of IE 7 except
tabbed browsing - and while it didn't invent tabbed browsing it did
perfect it.
>> Why don't you enlighten me on exactly what you think happened?
>>
> Ask Intergraph and ComputerVision how a reliance on UNIX's "superior OS"
> worked out in the long run!
Ah, the Artful Dodger ignores the question once again. Come on, a short
paragraph shouldn't be too hard for you...
--
Sincerely,
Ray Ingles (313) 227-2317
"If you have nothing to hide, you have nothing to fear."
- Slogan of the NKVD (forerunner of the KGB)
Edwards
> milling around and trying to clone the commercial products
You've claimed that before, a bunch of times actually. A very small
fraction of those times I've asked you what commercial product
TeX/LaTeX is supposed to be a clone of. You've never answered that
question (i.e. by naming an actual _product_, you've certainly
_replied_ at great length), and since I happen to be in that kind of a
mood, I'll ask again (not that I expect an answer, ha ha): what
commercial product do you imagine TeX/LaTeX to be a clone of?
> or at least add the popular characteristics as their vision of what
> is hot and what is not gets clearer with time.
For example? I mean, what "hot, popular characteristics" of, say, MS
Word are the TeX/LaTeX developers despeately trying to copy? (Ooh, I
know, is it superior mathematical formatting and typesetting?
Bibliography management? I can't wait to find out, this is so
exciting!)
--
Darrin