|
Linux Here, Linux There, Linux Everywhere |
|
Wednesday, January 07 2004 @ 09:30 PM EST
|
The Inquirer has a leaked internal IBM memo, they say from IBM CIO Bob Greenberg, asking all IBM executives to switch their desktops to Linux by the end of next year. After they do it, don't you suppose that will be the end of FUD along the lines of "Linux isn't ready for the desktop"? Everyone will just know that if IBM runs Linux on the desktop, so can any other business. Any more FUD of that nature will be answered immediately and reflexively in everyone's mind with: "But IBM uses it just fine." That's nice. One more piece of FUD will fall down and go boom.
SiliconValley.com has the story that Red Flag Linux and Japan's Miracle Linux are going to work together to create a unified Linux for China, Japan and Korea. They'll work together on the development, but each company will have it's own branded distro. Their motive? The article says it's their hope to challenge Microsoft's dominance: China's Red Flag Software Co. and Japan's Miracle Linux Corp. said they expect to complete the new system, called Asianux, by May.
"We want to have a unified platform throughout Asia," said Miracle Linux President Takeshi Sato at a news conference. Linux is an open-source system, meaning the code on which it is based is freely available to users who can contribute improvements. By contrast, Microsoft's source code is a company secret. Some Asian governments are uneasy that Windows might contain features that would allow outsiders to spy on its users. They also dislike having to rely so heavily on a single foreign company for the development of their technology industries.
Here's the part that caught my eye: Oracle Corp. owns 58.5 percent of Miracle Linux. Thinking of Switching? If you are thinking of switching to GNU/Linux you might find this web site helpful, "How to Evaluate Open Source Software/Free Software (OSS/FS) Programs" by David Wheeler. IBM has plenty of helpful information also, but they should, in my opinion, put a link to Linux info on their home page. Microsoft is also providing some "information" on Linux with its "Get the Facts on Windows and Linux" campaign. Their use of English gives a gal chills, with its Newspeak overtones. "Facts." Are you sure? It seems a bunch of "leading companies and third-party analysts" have determined that Windows has a lower cost of ownership and "performs better" than Linux. Lower than free? Microsoft finds analysts and companies willing to say they are better, but how could they stretch things around to find lower than free? If I thought anyone here took what Microsoft says about Linux to heart, I'd bother to debunk it all, but we've cut our teeth on the likes of Laura DiDio, so we're a hard sell when it comes to the opinions of analysts. If you wish to examine the list of their analysts and what they say they found, it is here. The Industry Case Studies page is here. To help you evaluate if Microsoft is providing facts or PR, here's an article in CIO magazine from September, which evaluated one of the reports, the one from "independent analyst" Giga Research, and also explained in the commentary why the author thought Microsoft commissioned Giga to do the study when it did: Of course, it’s not shocking that a study commissioned by Microsoft should demonstrate the advantages of that company’s products over Linux, but the fact that the study was commissioned at all is revealing of the big company’s concern. The popularity of Linux—fueled by fear of placing too much control in the hands of a single (notoriously aggressive) vendor and by the widespread conviction that open source software can save you a bucket of money—is rising like the waters of the flood toward the software fortress that Gates built. IDC, a sister company to CXO Media, recently reported that sales of Linux servers now exceed those of Windows servers, and Gartner tells us that while the sale of servers running Windows is up 30 percent this year over last, the sale of Intel servers running Linux is up nearly 60 percent in that same time frame. In short, it’s a very good time for Bill Gates to pull out his checkbook and order up some market research. How did Giga reach the conclusion that Linux, which is free, costs more than Microsoft? By comparing MS with high-priced Linux distros: The Giga study found that the biggest cost advantages of Microsoft products came in comparison to the cost of Linux-based products sold by monster software makers Oracle and BEA. According to the study, large corporations paid $80,000 for Oracle’s database, compared to less than $40,000 for Microsoft SQL; and they paid $60,000 to BEA for development tools, compared to $12,500 to Microsoft VisualStudio.net. Medium-size companies, the study found, enjoyed a savings of the same proportions.
What would the cost savings look like if the companies that paid big bucks to Oracle and BEA had used free Linux-based databases and scripting tools such as PHP and MySQL? Giga doesn’t know, because, as Cormier explains, they didn't look at any such companies. It's a crying shame to have to be so cynical. But a girl has to do what a girl has to do. You can't just believe everything you read. If Microsoft would like to know what else I like about Linux, here's something: I like it that Linus does his level best to tell the truth and that, no matter how much I dig, I never come up with anything slick, like these "facts" on the "lower cost of ownership".
|
|
Authored by: Alizarin on Wednesday, January 07 2004 @ 10:57 PM EST |
IBM moving to Linux for desktop use is a Good Thing(tm). It means that all
problems concerning ease of use, administration, etc in a corporate environment
will be addressed and, in all probability, released back to the community. This
should make Linux a stronger overall product.
On a side note, I'm considering moving over to Linux when I build a new dual
Opteron workstation in a few months :-)[ Reply to This | # ]
|
- Good for IBM! - Authored by: bobn on Wednesday, January 07 2004 @ 11:01 PM EST
- Good for IBM! - Authored by: Sunny Penguin on Wednesday, January 07 2004 @ 11:15 PM EST
- Desktop by end of 2004? - Authored by: Anonymous on Thursday, January 08 2004 @ 12:09 AM EST
- Why wait ?!? - Authored by: Anonymous on Thursday, January 08 2004 @ 12:39 AM EST
- Why wait ?!? - Authored by: Anonymous on Thursday, January 08 2004 @ 02:15 PM EST
- Good for IBM! Yes but... - Authored by: jaydee on Thursday, January 08 2004 @ 04:19 AM EST
- Good for IBM! - Authored by: Anonymous on Thursday, January 08 2004 @ 10:27 AM EST
- Good for IBM! - Authored by: Anonymous on Thursday, January 08 2004 @ 02:08 PM EST
- Hello, IBM guy! - Authored by: Anonymous on Friday, January 09 2004 @ 02:15 AM EST
- Hello, IBM guy! - Authored by: ihawk on Friday, January 09 2004 @ 12:55 PM EST
- Laptop + Linux - Authored by: Anonymous on Friday, January 09 2004 @ 03:13 PM EST
- Hello, IBM guy! - Authored by: Anonymous on Saturday, January 10 2004 @ 06:14 AM EST
|
Authored by: gnutechguy on Wednesday, January 07 2004 @ 11:00 PM EST |
The Microsoft site "Get the Facts" should be entitled "Get the
FUD." Microsoft will simply find compliant lackey analysts to serve up the
results they want. What the IBM move away from a non-Microsoft controlled
desktop should say to businesses is:
"If a Fortune 50 company can escape from Microsoft, then I should be able
also."
I hope this may start the herd away from Microsoft desktops.
I also try to remember that with Microsoft (and SCO) that truth is merely a
mallable tool for acquiring power.
Just one man's opinion.[ Reply to This | # ]
|
|
Authored by: hairball on Wednesday, January 07 2004 @ 11:05 PM EST |
Hi all,
The momentum is growing real fast now. Almost frighteningly so. The number of
fronts Microsoft has to fight on is getting out of control. Their target
audience are becoming more and more cynical.
Their TCO arguments are meaningless - you CANNOT own a microsoft operating
system - ever. Use it and you are NEVER in control of your costs, your
information, or your business. In fact we should be disputing Microsoft's use
of the phrase - misleading advertising it is - yes!.
The SCO fiasco has clearly demonstrated what happens when a company goes bad.
What will happen if Microsoft goes the same way? I do not want to be running my
business on their software if this happens.
For the last year or so we have been drumming into our IT students that if you
master Microsoft products you make money for them little for your client and
almost none for yourselves.
If you master the Open Source products you make money for yourself and your
client. The learning curve is steeper but the rewards stay within your local
community.
---
Cheers,
Hairball Lightspeed
From Here to Eternity in 15 seconds.[ Reply to This | # ]
|
|
Authored by: Anonymous on Wednesday, January 07 2004 @ 11:06 PM EST |
PJ, I love* and respect you, but how is this any better than a printing a
rumor or paraphrasing a press release? Since IBM did not apparently release this
memo for public knowledge, I thought you would leave this kind of speculation to
slashdot, where nothing that is said is really relevant.
Many apologies,
and I most definitely am not trying to tell you what to do with this
site.
*I also love freedom. You know what I mean.[ Reply to This | # ]
|
- Read the ENTIRE article - Authored by: Anonymous on Wednesday, January 07 2004 @ 11:14 PM EST
- Plea - Authored by: Anonymous on Wednesday, January 07 2004 @ 11:22 PM EST
- Plea - Authored by: Anonymous on Thursday, January 08 2004 @ 07:47 AM EST
- Plea - Authored by: whitehat on Wednesday, January 07 2004 @ 11:27 PM EST
- Plea - Authored by: Anonymous on Thursday, January 08 2004 @ 12:02 AM EST
- Plea - Authored by: Anonymous on Thursday, January 08 2004 @ 12:12 AM EST
- Plea - Authored by: whitehat on Thursday, January 08 2004 @ 10:36 AM EST
- And you are...whom? - Authored by: darthaggie on Thursday, January 08 2004 @ 11:24 AM EST
- Plea - Authored by: Anonymous on Thursday, January 08 2004 @ 02:56 PM EST
- Plea - Authored by: Anonymous on Thursday, January 08 2004 @ 10:12 AM EST
- Plea - Authored by: k12linux on Thursday, January 08 2004 @ 02:59 PM EST
- Is it safe yet? - Authored by: Tim Ransom on Friday, January 09 2004 @ 03:43 AM EST
|
Authored by: timmy on Wednesday, January 07 2004 @ 11:16 PM EST |
What delicious irony. Bill Gates' greatest accomplishment was taking the
computing monopoly from IBM. Now IBM is working as hard as it can to destroy
the computing monopoly.[ Reply to This | # ]
|
|
Authored by: toolboxnz on Wednesday, January 07 2004 @ 11:17 PM EST |
...have determined that Windows has a lower cost of ownership and
"performs better" than Linux. Lower than free? Microsoft finds analysts and
companies willing to say they are better, but how could they stretch things
around to find lower than free?
I am a Linux advocate but will
freely admit that the total cost of ownership of Linux is not free. The
cost of purchasing and installing it may very well be free but there are still
administration costs in an business environment for supporting networks and
users, lost downtime etc. This is what the TCO is all about. However, it is
generally recognised (at least in the Linux community) that it's far easier to
manage and maintain Linux/Unix networks that it is Windows networks, so the TCO
should therefore as you'll need less support resources.
While I'm
harping on, you generally pay something for Linux anyway. The only ways you can
get an actual free copy that I can think of are as follows:
1) You have
a free internet connection (or stolen bandwidth eg at work) and download the
distro by FTP
2) You have a nice friend who gave you a copy on a CD they
burnt for you
3) You go to one of those Linux installfests and someone
installs it for you
4) Your nice friend comes around with their copy on
CD and installs it for you
Any other way does actually cost you money,
whether it's in download charges, monthly internet access, paying $5 from
some website to send you a copy on CD etc etc.
End of pedantic
rambling... [ Reply to This | # ]
|
- TCO is not just about the purchase price - Authored by: gnutechguy on Wednesday, January 07 2004 @ 11:32 PM EST
- TCO is not just about the purchase price - Authored by: eamacnaghten on Wednesday, January 07 2004 @ 11:54 PM EST
- TCO is not just about the purchase price - Authored by: mrhartwig on Thursday, January 08 2004 @ 12:04 AM EST
- MS SQL @ $40,000 - Authored by: Anonymous on Thursday, January 08 2004 @ 01:02 AM EST
- TCO is not just about the purchase price - Authored by: hhouwaard on Thursday, January 08 2004 @ 02:18 AM EST
- TCO is not just about the purchase price - Authored by: hairball on Thursday, January 08 2004 @ 06:55 AM EST
- An IT Manager writes.. - Authored by: jaydee on Thursday, January 08 2004 @ 07:02 AM EST
- TCO is not just about the purchase price - Authored by: Anonymous on Thursday, January 08 2004 @ 10:31 AM EST
- TCO is not just about the purchase price - Authored by: Anonymous on Thursday, January 08 2004 @ 12:46 PM EST
- TCO is not just about the purchase price - Authored by: k12linux on Thursday, January 08 2004 @ 03:10 PM EST
- TCO is not just about the purchase price - Authored by: Anonymous on Friday, January 09 2004 @ 07:47 AM EST
- TCO is not just about the purchase price - Authored by: raindog on Friday, January 09 2004 @ 11:02 AM EST
|
Authored by: freeio on Wednesday, January 07 2004 @ 11:24 PM EST |
Corporate conversions should be in many ways more possible than individual
conversions, if only because for most purposes, large corporations simply
mandate what all employees shall use. I am reminded of the Ernie Ball
Manufacturing Company conversion, which was inspired by a BSA audit. The CEO
simply said, "Let it be so!" and within a short period of time it
was so. Users of most corporate desktops use a limited subset of the available
tools, and most can do their entire jobs using generally available free
software.
There will of course be difficult cases for specialized software, but those are
the exception in most companies. In our photo business, we make extensive use
of Adobe products, i.e. Photoshop, Illustrator, and PageMaker, primarily. There
is also a beige box which handles the big Epson 7600 printer driver, which will
not yet run on free software. But everyone else and everything else runs just
fine of SuSE Linux Professional 9.0. Buy one kit (I like having the manuals and
the DVD-based installation) and then install on the dozen systems we have. It
all just works, on every system we have. So what is there not to like about
it?
What we face is that most companies and educators confuse "learning
computers" with "learning Microsoft or Adobe Products." The
classes are all based on those, with no alternatives given. It is as if nothing
else exists. It is not that many skills cannot be transferred elsewhere, but
when what a person learns is the cutest keyboard macros to automate one brand of
program, well of course that person will feel cheated when the new program does
not work just exactly like the one she/he learned.
Quite frankly, the conversion to linux for us was no more difficult that buying
a computer with XP installed, and having to learn the new quirks after using 98
for several years. Linux is no more of a shock than that, which is expected
when we buy new hardware anyway.
So what is the big deal? Well, they call linux a disruptive technology for a
reason. The disruption is primarily that we are leaving the world where
"No one ever got fired for running Microsoft," is no longer true.
It is a far larger shock for the CIO and IT department than it is for the users.
---
73 de w4ti[ Reply to This | # ]
|
|
Authored by: Anonymous on Wednesday, January 07 2004 @ 11:33 PM EST |
I wonder if only because this was after all an internal IBM memo, and might not
be as precisely worded as a public press release. F'rinstance the Gnome desktop
is arguably (though not accurately) "Linux Based", but I believe
runs on AIX, Solaris and possibly other Unixes as well. The memo mentions end of
2005 timeframe, but its difficult to fathom IBM completely replacing AIX with
Linux in that timeframe. Moving to a common desktop environment e.g. Gnome is a
different matter.
Ed[ Reply to This | # ]
|
|
Authored by: Anonymous on Wednesday, January 07 2004 @ 11:40 PM EST |
I predict that by 2006 Microsoft will have its own distribution of linux as well
as LongHorn. That way if they have to lose people migrating from windows.. at
least they can get support contracts instead of an entirely lost business.[ Reply to This | # ]
|
- Microsoft Linux - Authored by: whitehat on Thursday, January 08 2004 @ 12:02 AM EST
- Linux Here, Linux There, Linux Everywhere - Authored by: Anonymous on Thursday, January 08 2004 @ 12:03 AM EST
- Linux Here, Linux There, Linux Everywhere - Authored by: Anonymous on Thursday, January 08 2004 @ 01:21 AM EST
- Linux Here, Linux There, Linux Everywhere - Authored by: joe on Thursday, January 08 2004 @ 02:18 AM EST
- Sisyphus' Ski Slope - Authored by: overshoot on Thursday, January 08 2004 @ 08:17 AM EST
- Linux Here, Linux There, Linux Everywhere - Authored by: Jude on Thursday, January 08 2004 @ 09:42 AM EST
- Support from MS? - Authored by: Anonymous on Thursday, January 08 2004 @ 10:45 AM EST
- *nix on M$ - Authored by: Anonymous on Thursday, January 08 2004 @ 01:28 PM EST
- BSDs - Authored by: bruce_s on Friday, January 09 2004 @ 08:25 PM EST
|
Authored by: Anonymous on Wednesday, January 07 2004 @ 11:45 PM EST |
Since PJ is so thoughtful of what Linus has said in the past... note that he has
also said that LINUX will not be ready for the desktop until 2005.
I wonder what he meant by that?
And maybe it is true!
So - in the meantime, OpenOffice.org and othe open source software should be put
on all existing computers with Windows (at least until more apps are migrated to
the LINUX desktop). OpenOffice or StarOffice is the 1st step that anyone today
can make to free themselves from the ropes of Microsoft's bondage.
2005 is only a year away... can't wait![ Reply to This | # ]
|
|
Authored by: Anonymous on Thursday, January 08 2004 @ 12:23 AM EST |
Everyone seems to mention MySQL as the free database of choice.
I would suggest that, for most purposes where Oracle is being used, PostgreSQL
would be a better choice than MySQL. Other possibilities are two commercial
database packages that have been open sourced: SAP/DB, and Firebird (derived
from Borland's briefly open sourced Interbase).
MySQL was never designed originally to support the full range of database
features (constraints, transactions, stored procedures, etc). Although some of
these features have been added onto MySQL in recent versions, PostgreSQL and the
other two are more mature and feature-rich systems.
Still, MySQL is notoriously fast if all you want to do is read data out of a
simple database, with lots of records. The point is, it may be the fastest open
source database package, but it is not the most sophisticated. There's a
trade-off between speed and features, and other database packages may be more
appropriate for your business.[ Reply to This | # ]
|
- Postgresql - Authored by: jricher on Thursday, January 08 2004 @ 01:35 AM EST
- Postgresql - Authored by: Anonymous on Thursday, January 08 2004 @ 05:42 AM EST
- Postgresql - Authored by: Anonymous on Thursday, January 08 2004 @ 08:52 AM EST
|
Authored by: DaveC on Thursday, January 08 2004 @ 12:27 AM EST |
What is a good distro for me? I've got HPUX and Win2k at work and 98 and XP at
home but am fed up with my home computer OS and want to use linux finally. I
sure hope Doom3 comes out for linux. Then I can play and work/doddle on the same
OS. :-) What is taking linux so long to play windows games?[ Reply to This | # ]
|
- what's a good distro? - Authored by: eamacnaghten on Thursday, January 08 2004 @ 12:44 AM EST
- what's a good distro? - Authored by: Jimmyoh on Thursday, January 08 2004 @ 01:17 AM EST
- what's a good distro? - Authored by: swillden on Thursday, January 08 2004 @ 01:26 AM EST
- what's a good distro? - Authored by: Anonymous on Thursday, January 08 2004 @ 01:38 AM EST
- what's a good distro? - Authored by: Alastair on Thursday, January 08 2004 @ 04:46 AM EST
- Games... - Authored by: jesse on Thursday, January 08 2004 @ 10:01 AM EST
- What's a good distro? - Authored by: Anonymous on Thursday, January 08 2004 @ 09:55 AM EST
- what's a good distro? - Authored by: resst on Thursday, January 08 2004 @ 10:29 AM EST
- what's a good distro? - Authored by: Anonymous on Thursday, January 08 2004 @ 03:30 PM EST
- what's a good distro? - Authored by: vonbrand on Friday, January 09 2004 @ 01:53 PM EST
|
Authored by: eamacnaghten on Thursday, January 08 2004 @ 12:33 AM EST |
PJ - In your article you quoted...
"What would the cost savings look
like if the companies that paid big bucks to Oracle and BEA had used free
Linux-based databases and scripting tools such as PHP and MySQL? Giga doesn’t
know, because, as Cormier explains, they didn't look at any such
companies."
It is worth noting that both PHP and MySQL are available for
MS-Windows as opensource packages, as are Apache, Postgres, OpenOffice, Mozilla,
Perl and hosts of others including most of GNU, both under Cygwin and (though
not so extensibly) natively.
However, these packages are designed to run
under Linux/UNix environments, and do so better there. Also there is the
constant fear that MS will put a "feature" in the next version of the OS that
will prevent these from running.
Where Linux scores over MS-Windows over the
TCO is the ability to automate and schedule tasks easier due to the
accessibility of the internals, and the prolification of tools created by others
to help you do so. In short, in my experience, you tend to spend far fewer
hours looking after Linux boxes than you do MS-Windows ones. [ Reply to This | # ]
|
|
Authored by: Anonymous on Thursday, January 08 2004 @ 12:43 AM EST |
Also nice to notice when you visit M$'s site, are the comments posted by the
visitors, as the majority of them are so suspiciously pro-Microsoft. Well...
maybe they're not just visitors, but... Yes! They are part of M$'s anti-Linux
campaign too. Instead of fixing their ill behaved products, M$ employees have
been given the task of attending as many forums as they can, in order to give
their wise advice, plenty of 'facts', and other FUD. [ Reply to This | # ]
|
|
Authored by: whoever57 on Thursday, January 08 2004 @ 12:56 AM EST |
I recall that it was probably about 18 months ago ago that the press was saying
the Linux was not yet ready to be used as a server. There were many articles
saying htat Linux had "not proved itself". This was despite several
high-profile Linux-based installations, such as Google, Amazon, etc.
Of course, Linux was ready and those who were technology leaders, not followers,
were already well down the road of deployment.
It seems to me that we are just getting to the same point with Linux on the
desktop. The tipping point has already been reached, yet the followers (and
those who don't want to see it) don't accept the fact yet.
My guess is that in a year or two, those early adopters such as IBM and Munich,
will be seen as merely part of the mainstream.
---
-----
For a few laughs, see "Simon's Comic Online Source" at
http://scosource.com/index.html[ Reply to This | # ]
|
|
Authored by: davcefai on Thursday, January 08 2004 @ 01:16 AM EST |
After trying out Linux at home every 12-18 months for about 8 years, last year I
decided that Mandrake 9.0 was usable. In the past 7 months I have gone from
total newbie to "just coping", but can see my way to "expert
user" status.
At work we are slowly switching from MS Office to StarOffice (still on Windows).
We could just as easily have gone for OpenOffice but wanted Sun's backup.
With this background:
1. I agree that FOSS software is capital cost free - you cannot get any
cheaper.
2. The learning curve is VERY steep, to the extent that, initially, productivity
can drop.
3. Most users are hideously conservative and will oppose change on principle.
4. So are their bosses. You need a dictum from "on high" to override
their objections.
5. Once you're over the initial hump(s) you're laughing all the way to the
bank.
The way I see it, once you've dented the MS monopoly, the process will
accelerate. The newer versions of Linux - Mandrake 9.x and Sun's Java Desktop
(now that's a silly name) will help the process tremendously.
No matter how much FUD the MS camp generates, it cannot compete against the
succes stories and a cold blooded, properly performed assessment of FOSS.
The biggest handicap to Linux desktop adoption is the lack of suitable
replacements for certain Windows Programs. MS Access, Autocad, Paint Shop pro
spring to mind.
[ Reply to This | # ]
|
|
Authored by: Anonymous on Thursday, January 08 2004 @ 01:22 AM EST |
<p>PJ, did you not how you used the word "free" in the above
article? Right, it's "free", as in beer, as opposed to
"free" as a human right. If we use the word that way, we'll fail in
two ways:</p>
<ul>
<li>We do identify the wrong source for Linux' success.
Of course, having not to pay for Linux helped me to
use it privately. However, to prefer it in my business
use (where I am <em>not</em> paying the software I use)
it was much more important to have the source,
being able to fix errors, add features and all that
stuff.</li>
<li>We may easily be contradicted with the simple phrase
"TCO". And, believe me, there is such a thing like
"TCO". It is not that easy to explain, but I try
it this way: Consider preparing some Pasta for you
and a friend. Nice thing, easy to do, delicious
result. Works fine. Now consider preparing the same
pasta for a 20 people. You'll soon find, that it
becomes a little bit difficult. The water temperatur
differs a lot more and it is diffucult to get it hot
again. You have no feeling for how much salt to use.
And now consider preparing the same pasta for
50000 people. (These are the numbers in which
corporate administrators may very welll think.)
It's a nightmare.<br>
In other words: It may finally be cheaper to
pay for professional services, additional tools,
and the like. And the factual costs are the
sum of all costs, not just the licenses.</li>
</ul>
[ Reply to This | # ]
|
- "Free" as in beer - Authored by: Anonymous on Thursday, January 08 2004 @ 11:44 AM EST
|
Authored by: Joss the Red on Thursday, January 08 2004 @ 01:41 AM EST |
Am I the only one that finds "Performs better" tougher to gag down
than "Lower total cost of ownership"?
---
use IANAL.pm[ Reply to This | # ]
|
|
Authored by: Anonymous on Thursday, January 08 2004 @ 02:03 AM EST |
Well, I bothered to waste an hour and look through the
volumes of pages
on these tests, and I came across
something that I find quite notable:
This link covers a benchmark study
conducted by
VeriTest, and what's interesting is that if
you go to page 14 of this
PDF report, there's a table that
contains some damning information
regarding these
benchmarks. Here is the table that describes the file
system parameters:
Operating System
........
Block Size
Windows Server 2003
........
64k
Red Hat Advanced Linux Server 2.1 ........
default
Now most techies know that block size on
the file
system really affects disk I/O performance. There is a
trade-off
when choosing a larger block size, versus a
smaller block size. The larger
block size gives faster
throughput, at the expense of less usable disk space
(this
typically doesn't matter now since disk drives are huge).
The
smaller the block size, the slower the throughput, but
gives more usable
space. Why do you suppose they didn't
cite the Linux file system block
size???
Quick example. If I choose 32k block size file
system,
and write a file that is 2k in size, the file system will
still use
up 32k. If you choose 4k block size, and write
a 12k file, 12k is used, no
more no less (3 blocks used).
While the latter optimizes disk space usage, it
also slows
down the I/O.
Knowing all of this, can someone
please cite what the
default block size is on Linux install? (I can't seem
to
find this in my SuSE distro.)
JMonroy [ Reply to This | # ]
|
- Apples to apples, or apples to oranges? - Authored by: Alex on Thursday, January 08 2004 @ 02:11 AM EST
- Apples to apples, or apples to oranges? - Authored by: Anonymous on Thursday, January 08 2004 @ 02:24 AM EST
- Apples to apples, or apples to oranges? - Authored by: Anonymous on Thursday, January 08 2004 @ 12:11 PM EST
- Apples to apples, or apples to oranges? - Authored by: JMonroy on Thursday, January 08 2004 @ 12:53 PM EST
- Apples to apples, or apples to oranges? - Authored by: Anonymous on Thursday, January 08 2004 @ 02:22 PM EST
- Clickable link here - Authored by: Chugiak on Thursday, January 08 2004 @ 05:46 PM EST
|
Authored by: whitehat on Thursday, January 08 2004 @ 02:16 AM EST |
In the heat of the battle with SCO, we sometimes for get the
Halloween
memos, a series of memorandums issued from Redmond re:
the Linux threat. For
newcomers to Linux, here is the link Halloween Memos
These memos also shed some light on the situation
with SCO. In
Microsoft's analysis, Linux is inevitably taking over from
proprietary Unix
platforms. In Microsoft's eyes, SCO is already an
irrelevant technology
player.
Caveat: Microsoft has pointed
out, that employees write
memorandums that do not necessarily
reflect the company's official position. [ Reply to This | # ]
|
|
Authored by: Anonymous on Thursday, January 08 2004 @ 02:39 AM EST |
This one is official, though. You can
hire IBM to help you, and you can watch what IBM does for itself. [ Reply to This | # ]
|
|
Authored by: Anonymous on Thursday, January 08 2004 @ 04:58 AM EST |
You quote (I've snipped it a little) "According to the study, large
corporations paid $80,000 for Oracle’s database, compared to less than $40,000
for Microsoft SQL; and they paid $60,000 to BEA for development tools, compared
to $12,500 to Microsoft VisualStudio.net."
As a developer of large enterprise systems, it seems to me this is not comparing
like with like. Let me give you a comparable quote I just made up :
"According to the study, large corporations paid $300,000 for large
commercial trucks, compared to less than $20,000 for passenger vehicles."
The point I'm making is that Oracle and BEA are tools you'd spend a lot of
money on for large systems, where small-system tools like MS SQL Server
wouldn't handle the load. These large-system tools cost more than small system
tools. Is anyone suprised at that?
If someone told you that a truck cost 15 times more than a car would you buy the
truck or the car? Well, it depends on the job. To move one person you'd use the
car, though the truck could still do it. But to move 30 tons 1000 kilometres
overnight you'd need the truck. The car wouldn't do the job, even if it was
cheaper.
So comparing just on cost is ludicruous. You have to look at what the more
expensive system is being used for.
[ Reply to This | # ]
|
|
Authored by: RSC on Thursday, January 08 2004 @ 05:04 AM EST |
This bit about IBM could make my day.
The only thing stopping me from using Linux on my work desktop is Lotus NOTES. I
hope the the move to linux within IBM is an indication that they will finally
release a linux NOTES client.
RSC
---
----
An Australian who IS interested.[ Reply to This | # ]
|
- Linux Here, Linux There, Linux Everywhere - Authored by: Anonymous on Thursday, January 08 2004 @ 05:55 AM EST
- Bloated Goats - Authored by: overshoot on Thursday, January 08 2004 @ 08:44 AM EST
- Lost Notes - Authored by: Anonymous on Thursday, January 08 2004 @ 11:34 AM EST
- Native Linux Notes client unlikely imho - Authored by: mnuttall on Thursday, January 08 2004 @ 09:54 AM EST
- It's more than just eMail - Authored by: Anonymous on Thursday, January 08 2004 @ 03:22 PM EST
- Linux Here, Linux There, Linux Everywhere - Authored by: Anonymous on Friday, January 09 2004 @ 08:10 AM EST
|
Authored by: Anonymous on Thursday, January 08 2004 @ 05:42 AM EST |
M$ seem to have found the courage to tackle Linux head on with facts and
figures. (albeit heavily massaged ones). This seems like a major initiative.
We should embrace this opportunity to war on facts.
Eventually M$ can only tangle themselves in the web of FUD they spin.
The fact that they are doing this shows that they are 'concerned' enough about
linux, they cannot ignore it any longer.
In a war of public relations, getting your adversary to acknowledge your
presence as serious competition is the first battle.
Should be downhill from now.
[ Reply to This | # ]
|
|
Authored by: Anonymous on Thursday, January 08 2004 @ 06:08 AM EST |
Obviously IBM is not planning to have to pay SCO licencing fees !!
[ Reply to This | # ]
|
|
Authored by: TobiasBXL on Thursday, January 08 2004 @ 06:14 AM EST |
I don't get the attention this Microsoft campaign is getting. I mean, those
"studies" have been around for almost a year, take the one from
Gartner for example.
Every person in the business knows those studies are principally nothing new and
most of them have been paid for by Microsoft. Do you go to Ford to inform
yourself if you want to buy a car from Chrysler?!
Have a look at this article which picks up on the paid studies idea:
http://www.eweek.com/article2/0,4149,1426252,00.asp
Some of the highlights are:
"But Taylor's [a Microserv] plan to provide customers with
"objective third-party research and facts," much of which is paid
for or sponsored by Microsoft itself, has not always been that
well-received."
"... a research study that was paid for by Microsoft..."
and in the same context
"...the research firm later said it would no longer publicize any similar
future studies."
I like this one too:
"...actively been lobbying governments around the world..."
because we all know what "actively lobbying" means:
"Dear mayor, in one hand I am holding a contract worth millions in
software sales and in the other I am holding a very large bundle of money. If
you'll sign the contract I really wouldn't notice the absence of that
money." :-)
Looking at it relaxed though, this MS campaign is going to do the opposite of
what MS intended. They really should know. We have all read that Halloween
document which describes the horror of people discussing in an MS memo that
their FUD has been backfiring into their faces. This will happen again.
cheers,
Tobias[ Reply to This | # ]
|
|
Authored by: Steve Martin on Thursday, January 08 2004 @ 06:56 AM EST |
This has got to frost McBride and Co. severely...
IBM is
about to compete head-to-head with SCO's point-of-sale systems, using
Linux-based boxes. Not just any Linux distro, either: (drum roll, please)
They're gonna use SuSE.
Here
's the Forbes article, from Reuters.
--- "When I say something, I
put my name next to it." -- Isaac Jaffee, "Sports Night" [ Reply to This | # ]
|
|
Authored by: Anonymous on Thursday, January 08 2004 @ 06:59 AM EST |
Total Cost of Ownership (TCO) is a very vague term and I see it being bent to
suit particular interests all the time. If you never deploy or use the software
then a Linux distro will almost always have a lower TCO than any MS Windows
product(s) with similar functionality. Simply because then TCO is the purchase
price and nothing else. However, that is not to say that Linux is free, and why
should it be? Since when did "free" in Free Software imply gratis?!
I have no statistics on the matter so I'll take a few personal examples.
At the University where I work (as a researcher) the mail-serves were diabled
(due to high load) when SobigF was most active, leaving thousands of researches
without e-mail for a couple of days. What is the cost of that? Locally we use
Macs and a few Linux PCs which were obviously immune, but we still couldn't use
e-mail. Likewise, individual Blaster infected computers had to be blocked from
accessing the intranet by dynamic reconfiguration of network switches. What is
the cost of that? Microsoft's line of reasoning is that had we all patched
those thousands of Windows machines the costs would have been zero, but that
doesn't change the fact that the costs were not zero. Shouldn't those costs be
included in the TCO?! And please remember that educating thousands of people on
using Windows Update on a daily basis is also not without costs, albeit they may
very well be lower.
A major ticket reseller here in Denmark had to shut down operations for several
days due to problems with their SQL server which held the booking data. You
simply couldn't order tickets and I know that they wrote off close to $200000
per day. Shouldn't that downtime be taken into account when calculating the
TCO?!
Where I live we have a small network with 60+ computers to give cheap internet
access to all the residents. I used to administer our Linux firewall. It
survived several power cuts, though it ran Red Hat Linux 6.2 which didn't come
with a journalling filesystem out of the box. However, when we changed to a MS
Windows 2000 Server (which I don't admin) it only took a single power cut to
require a complete reinstall. That left more than 60 people without internet
access for five days. Yes, we should've bought an Uninterruptable Power Supply
(UPS), but we didn't have one. Again, should this downtime count against the
TCO?!
My point is that there are so many factors that go into the real world TCO that
the actual number is useless unless you know all the details of the analysis -
and that they are identical when you compare different studies. Also there seems
to be quite a difference between how systems *should* be deployed and how they
are in fact deployed in the real world.
And I haven't even touched upon my personal experience that linux admins are
often much more knowledgeable than windows admins, even when you hire them for
the same salary. Should that go into the TCO as well, and how?!
[ Reply to This | # ]
|
|
Authored by: Anonymous on Thursday, January 08 2004 @ 07:04 AM EST |
Pamela,
I have a friend who works at IBM and he told me 3 or 4 months ago that IBM WILL
be moving to Linux on all of its desktops in the next year. Not a maybe, but a
WILL. IBM is already training relevant staff on Linux deployment from what he
indicated. Of course, I have no 'solid' proof of this, other than my friends
word, but he's not the type to lie. This 'leaked memo' seems to confirm what
he has told me.
Dave[ Reply to This | # ]
|
|
Authored by: PeteS on Thursday, January 08 2004 @ 07:09 AM EST |
Well, let's all keep in mind that M$ faces stiff competition in the academic
world too, where many times the computer of choice is Apple
Although the desktop is where M$
is ubiquitous, servers are where the big margins are (and is a faster growing
segment of the market anyway), and now Virgina Tech builds
world's #3 supercomputer with Apple G5s
The people using this (it's in
the college of engineering) are the next generation of engineers and admins, and
they are cutting their teeth on FOSS (because the InfiniBand stack that they can
freely access is a sourceforge project. This is the interconnect that allows
them to get supercomputing performance from lots of workstation class machines,
at a fraction of the historical cost
As a disclaimer, I work in that
technology area, but it's interesting that an open standard is driving the cost
of supercomputing down (significantly) and runs on FOSS Operating
Systems.
As an aside, the primary vendors of supercomputing interconnects
use closed, proprietary hardware and software with very large margins. An open
standard has forced them to compete; a similar situation to which M$ now is
starting to find itself.
So let's not forget that there are many markets
where Microsoft, and indeed many companies using the proprietary, closed model,
is not only being engaged, but outdone completely, based on open standards, FOSS
systems and utility hardware.
--- Artificial Intelligence is no match
for natural stupidity [ Reply to This | # ]
|
|
Authored by: Anonymous on Thursday, January 08 2004 @ 07:39 AM EST |
I like reading Groklaw and have never posted but reading this part -
What would the cost savings look like if the companies that paid big bucks to
Oracle and BEA had used free Linux-based databases and scripting tools such as
PHP and MySQL?
- I had to reply
MySQL and PHP, in their current incarnations, could never hope to replace
Oracle's Database and BEA's Application server.
The forthcoming version 5 of mySQL will have some new interesting features may
bring it into the scope of dapartmental datamarts, but for bigger DB's you
really need SQLServer, 10g or DB2. And for the biggest ones only 10g or DB2
will cut the mustard.
Timbo.[ Reply to This | # ]
|
|
Authored by: lnx4me on Thursday, January 08 2004 @ 07:44 AM EST |
the fire, you can check here for a BSD
flavor for a desk-top.
I prefer SuSE myself, it just seems as if SuSE
took the time to 'get it right'.
OT: Chris DiBono (I think the spelling
is correct) compiled the 2.6 kernel on Tech-TV last night. I observed; am I now
'tainted'?
Bob[ Reply to This | # ]
|
|
Authored by: Anonymous on Thursday, January 08 2004 @ 07:49 AM EST |
The Inquirer has posted linux stories before and in general they leave out a lot
of facts to make the story very one sided. Like the time they basically accused
any business that uses linux and doesn't have programmers activly adding to the
kernel stealing linux. Not to say this story isn't true, but they might just be
reporting only what they want to report.[ Reply to This | # ]
|
|
Authored by: overshoot on Thursday, January 08 2004 @ 07:53 AM EST |
Seeing as how I spent a good bit of yesterday putting together the flagship
Linux engineering workstation to replace our aging Suns. It says a bit about
the utility of Microsoft platforms in our environment when you have engineers
using every trick in the book to keep IT from replacing their seven-year-old 350
MHz Sun boxen with new Dell machines at eight times the clock speed. [ Reply to This | # ]
|
|
Authored by: Anonymous on Thursday, January 08 2004 @ 08:16 AM EST |
IBM aren't the only one...
Can you guess the name of the networking solutions company that is not renewing
their corporate M$ licenses this year and is moving all their staff over to
Linux and Openoffice ?
Here's a hint: They own their own Linux distro :)
[ Reply to This | # ]
|
|
Authored by: photocrimes on Thursday, January 08 2004 @ 08:41 AM EST |
Speaking of Microsoft FUD, about a year ago I had researched what they then had
claimed was a "Migration from Linux to Windows" for some small ISP
(I forget the name now, at the time it was the only one they had listed as a
success story). The story was full of how easy it was and how much money they
were going to save. If anyone remembers or can find it, please post it. I can no
longer find it in my records.
Well, here is the funny part. During my investigation I discovered that the
hosting company was only running their own company website on Windows 2000. ALL
of their hosted sites, 3 years after the "successful" migration were
still being hosted on fresh up to date RedHad Linux servers. In fact, out of
about 25 seperate machines I looked into guess how many were Windows based?
"2" yep just 2. In 3 years they managed to migrate 2 machines. One
was their webserver and the other I could not get an aswer on. I then dug into
the technical forums that some of their employees fequented. It's funny, they
didn't seem to share Microsoft's idea as to what the word
"success" means. The only thing I could find that could be declared
a kind word in Microsoft's favor was one of their techs stating the webserver
was pretty easy to configure but hell to customize.
Needless to say, as of the time I was looking into it, the results did not match
what Microsoft claimed and a bulk of the work 3 years after the
"migration" was still being handled by the trusty Linux survers. New
up to date ones at that.
Microsoft doesn't want you to research their claims and in fact counts on most
of you not researching. Proprietary companies who rely on FUD as a sales method
must hate this site.
---
//A picture is worth a thousand words//[ Reply to This | # ]
|
|
Authored by: Anonymous on Thursday, January 08 2004 @ 08:47 AM EST |
A few quick observations:
1. IBM Moved their desktops and laptops from OS/2 to Win95 back in 1996/1997.
2. IBM's staff runs a lot on laptops nowadays. I hope they replace Windows
with Linux on their ThinkPads also. This would make it easier for the world to
do the same. I run Mandrake on my ThinkPad X23 and could use a few
additional/better drivers.
3. Many people at IBM use Lotus SmartSuite (Lotus Word, 123, Freelance,
Approach, etc). I doubt IBM/Lotus will port the suite to Linux. What's going
to happen to that product if IBM no longer uses it? How happy will the Lotus
people be? Will IBM write OOo import filters to read SmartSuite's file
formats?
4. The day of a native Notes client for Linux may be near. That'll make a lot
of people happy!
5. IBM's staff (and mainframe or AS/400 customers) use a 3270/5250 emulator
called IBM Personnal Communications (PCOMM), which only runs on Windows and
OS/2. Will IBM port PCOMM to Linux, or will they add SSL support and other
goodies to x3270? Either way, I'd be happy!
[ Reply to This | # ]
|
|
Authored by: linuxbikr on Thursday, January 08 2004 @ 08:53 AM EST |
"According to the study, large corporations paid $80,000 for Oracle’s
database, compared to less than $40,000
for Microsoft SQL; and they paid
$60,000 to BEA for development tools, compared to $12,500 to Microsoft
VisualStudio.net."
Some thoughts on these numbers...
- Oracle
uses a pricing model that can best be described as "customer gross revenue
percentage" + (hardware * # of CPUs). I have seen Oracle charge $5000 to
one small business customer for their product and charge $50,000 to a large
company for the EXACT same product on the same hardware! The reason: The large
company had ten times the revenue, hence could afford the higher cost. My
experiences with Oracle indicates they typical engage in practices that would be
charitibly described as highway robbery. SQL Server tends to be licensed on a
fixed cost * # of clients basis, hence the lower cost. Oracle will ask a lot of
questions about the intended use of the server and then determine a price. You
can pretty much buy SQL Server off-the-shelf and install it based on the number
of clients you want to talk to it independent of the hardware it is running
on.
- Comparing SQL Server to Oracle is unfair. You will never see a SQL
server running truly huge databases. That is the realm of DB/2 and Oracle.
Why? Because SQL Server is legendary for its special quirkness and instability
under large loads and it does not scale well. You'll see SQL Server a lot in
the low and middle tiers and you'll see overlaps in the middle tier with Oracle.
Once you reach into massive workloads, SQL Server goes down in flames. I use
Oracle a lot (but not fond of it) and I've seen it take abuse that would destroy
SQL Server. You'll never see SQL Server running a bank or insurance company
(and Oracle is hit or miss in that arena as well at times) but you'll see a lot
of IBM.
- BEA development tools usually include an application server.
It is very rare to buy BEA-specific tools without the server to run the
resulting products on. Having worked with BEA products in the past, I can say
with that that server license is part of the cost and it isn't cheap
($15K-17K for a J2EE clusterable server license). Plus, you will also need
to pay smaller fees to get keys to unlock development versions of the servers to
test the software on. I don't believe that Visual Studio ships with development
versions of the various servers needed to run the end result
application.
- Comparing Visual Studio to BEA products is deceptive. BEA
products are Java based and the high-end J2EE tools are pricey. I can get J2EE
development tools for FREE (Eclipse being one example) that can do many of the
same tasks as the BEA suites and still be able to use BEA-specific APIs. BEA
aside, there are lots of ways to write J2EE applications that don't require huge
monetary outlays or expensive product suites (I've done a lot of commercial J2EE
web development that had a zero dollar cost for the tools to my company but
deployed on $15,000 per server BEA application servers).
- A recent
database shootoff showed that MySQL 4.0 beat everyone for departmental and web
related workloads except at the very top end where Oracle nudged it out by a
hair. MySQL costs $300 for an unlimited user, unlimited hardware, "abuse it
any way you want" commercial license. Or you can use the free version and pay
nothing (I would pay for the commercial license for some extra features and to
support MySQL development if I deployed it commercially). Compare that price
tag to the number above and having any language or product talking to MySQL is a
win-win situation. I've lobbied unsuccessfully to have organizations bring in
MySQL servers on Windows or Linux rather than Oracle or SQL Server for databases
containing a million rows or less. It's dirt simple to set up (10 minutes on
Windows, a little longer on Linux for a binary install) and just screams. For
the cost of a single SQL Server license, you can buy a couple of very nice
Linux/Windows boxes to run your database on and have redunancy and never have to
pay another license or maintenance fee again.
[ Reply to This | # ]
|
|
Authored by: Anonymous on Thursday, January 08 2004 @ 09:37 AM EST |
Michael Overly, of Optimize Magazine, has an article posted by Yahoo! News
titled Protecting Against Open Source
Legal Risks that is at best cautiously against OSS. Yet another FUD
campaign, or is he that ignorant? He states that "This risk is
exacerbated by the nature of the open-source culture, which freely modifies and
distributes the software but without standard warranties, indemnities, and other
contractual protections." Yeah, and M$ protects us real
well. -David L. [ Reply to This | # ]
|
|
Authored by: zjimward on Thursday, January 08 2004 @ 10:26 AM EST |
I believe the whole desktop argument to be a matter of perspective. In the days
of DOS when Microsoft wanted to introduce Windows as the desktop many complained
about using the mouse, clicking icons and the list went on. In fact, I remember
it was the fact that you could just click and icon that was suppose to make it
so much easier than remembering and typing DOS commands. Funny, now it the
arguments installing software and pretty interfaces. Seems to me that Linux is a
good as Windows when compared to the simple fact that all that really is need is
simple access to software. In the Enterprise installation isn't left to the
average user. In small businesses there is always some one with computer
knowledge, either in or out of the company. We aren't talking rocket science
here. Every article I read that says Linux isn't ready for the desktop is a lot
of FUD. Basically the type that plays on the fear of the unknown. From my days
of consulting we were always told that no one likes change and Linux represents
that fear. I have used Linux for 7 years now and am happy for the change and the
choices that it has opened up for me.[ Reply to This | # ]
|
|
Authored by: Anonymous on Thursday, January 08 2004 @ 10:27 AM EST |
I've read some of those Get then "Facts" reports and there are so
many things wrong about them I feel like arguing it is like fighting with air.
Just to point out a few things. There are only totals for supposedly existing
systems serving supposedly the same number of users. No detailed breakdown of
costs, No application descriptions. No Bandwith/transacions per sec./application
size or name/State and age of hardware information.
That means that what they are saying is probably true to the facts they CHOSE to
include in their number crunching the same way we could mathematically prove
that the Earth is flat based on a CHOSEN set or undisputable facts.
I've said it before (although not here) and I'll say it again.
It's pointless to argue with people who can't hear you because their ears are
being covered by their butt cheeks...
---
No guns, no bombs...just brains
The way it should be.[ Reply to This | # ]
|
|
Authored by: blacklight on Thursday, January 08 2004 @ 10:42 AM EST |
The leaked IBM memo would be consistent with an IBM strategy of creating deep
inroads into the desktop market before Sun Microsystems' Mad Hatter achieves
any foothold in the desktop marketplace. Linux may or may not be ready for the
desktop market, but if Linux is not ready then it is awfully close to ready. By
mandating the changeover to Linux, IBM clearly wants to put that conversion into
its consulting achievements list. And IBM also wants its executives to think of
the future of IBM as closely tied to Linux and make their key decisions
accordingly.
If I were a Microsoft marketing drone right now, I would be paranoid. Linux has
penetrated and is penetrating into many if not most Microsoft installations. FUD
may be effective in preventing a competitor from gaining a foothold, but it's
useless when that competitor already has achieved its foothold. It is true that
it takes several years to master Linux but for many American corporations, the
Linux penetration started several years ago: at this moment, the Linux explosion
on both servers and desktop could happen at any time - all at the expense of
Microsoft's market share. Microsoft is going to be living in some interesting
times.[ Reply to This | # ]
|
|
Authored by: Anonymous on Thursday, January 08 2004 @ 10:54 AM EST |
Well, this is interesting, and is a significant step towards Linux becoming
widely used in the home. IRC, Xerox (or was it Verizon) switched their
development/technical staff over to Linux on the desktop. IBM doing the same
here (although on a corporate-wide level) will bring about some factors which
made the Wintel platform that standard for homes.
IBM's change to the Linux platform will cause other corporations which work w/
IBM to support interal installs of Linux (or at least OOo) for document exchange
/ compatibility issues. The initial installs are important, as this will create
momentum, as exposure is key to adoption (if it's as good as MS Office, has a
better price point, can be supported commercially, why not use it instead?)
As corporations move to Linux on the desktop, this exposes the average user to
the platform. Don't underestimate the work-at-home factor. This will cause
people to at the least examine Linux for their home use, if not switch. IMO
this was a major factor in the PC adoption rate in homes, the ability to perform
work tasks w/o needing to go to the office. The same force applies here,
although the costs are far less for the home user than an initial PC purchase.
Linux on the desktop might come quicker than a lot of people think. Change
usually does not happen quickly, but it seems as though IT is one are that moves
much more fluidly than traditional systems.
On a side note, just yesterday I finished building a Fedora box for my parents
(took a bit of time to find the drivers for the winmodem on Linux), and
installed it for them. They're far from power users, but their initial
reaction is positive, and I think they'll be far happier w/ this machine than
their previous one.
-Tomcat[ Reply to This | # ]
|
|
Authored by: Anonymous on Thursday, January 08 2004 @ 11:22 AM EST |
WAR IS PEACE!
FREEDOM IS SLAVERY!
LINUX COSTS MORE THAN WINDOWS (in the long run)![ Reply to This | # ]
|
|
Authored by: Anonymous on Thursday, January 08 2004 @ 11:27 AM EST |
I have worked at major IBM offices in the UK, Canada and other places, and they
were all using Win 98. Unless they have changed in the last couple of years they
have 2 options, now that MS has dropped 98 support: - upgrade everyone to
Win2K, or switch to Linux. If they switch to Win2K they will be paying hundreds
of millions to MS, at a time when MS are showing increasing aggression against
open source and indirectly IBM. If IBM do switch internally to Linux, I think
it will be a very beneficial move for them, and for the Linux community as well.
Perhaps we will finally see a Linux port of the Notes client as well (which
people have been clamoring for for years).
[ Reply to This | # ]
|
|
Authored by: Anonymous on Thursday, January 08 2004 @ 11:43 AM EST |
PJ,
"... but how could they stretch things around to find lower than
free?"
The main problem here is that they overpriced the cost of the migration and
forgot to mention that you have to change your Windows version each 2 years.
Yes, changing minds to use a different interface (no matter how little it
changed) could be tricky and costly, but you have to do it just one time. Change
Windows version each two years is much more costly if compared.
- "Look at how much you will pay now ? no, you don't need to look at
how much you will pay tomorrow."
The trick in those analyses is to use the best situation possible to Windows,
ever.
PR3J - Brasil[ Reply to This | # ]
|
|
Authored by: Thomas Frayne on Thursday, January 08 2004 @ 12:39 PM EST |
PJ, you said: "If you are thinking of switching to GNU/Linux you might
find this web site helpful, How to Evaluate Open Source
Software/Free Software (OSS/FS) Programs by David Wheeler. IBM has plenty
of helpful information also, but they should, in my opinion, put a link to Linux
info on their home page.
Microsoft is also providing some "information"
on Linux with its "Get the Facts on Windows and Linux" campaign.
"
However, you didn't mention that Wheeler provides a detailed
quantitative list and analysis of the facts that Microsoft misrepresented. I am
sure that Microsoft would like to ignore this article. We need to make it
widely known, and to make sure that follow-on studies get reported the same
way.
Wheeler's article is a definitive response to Microsoft's FUD. He
provides quantitative data over a multi-year period comparing Open Source
Software / Free Software (OSS/FS) to Microsoft's software, and shows that in all
respects OSS/FS is at least comparable to Microsoft's products, and in most
respects OSS/FS is a better choice. The article is liberally sprinkled with
links to the sources of all of the many studies he lists and analyzes, and the
total effect is overwhelming.
Here are his conclusions, with links to
the major sections of the article:
OSS/FS has significant market share
in many markets, is often the most reliable
software, and in many cases has the best performance.
OSS/FS
scales, both in problem size and project size. OSS/FS software often has far
better security,
perhaps due to the possibility of worldwide review. Total cost of ownership
for OSS/FS is often far less than proprietary software, especially as the number
of platforms increases. These statements are not merely opinions; these effects
can be shown quantitatively, using a wide variety of measures. This doesn’t even
consider
other issues that are hard to measure, such as freedom from control by a
single source, freedom from licensing management (with its accompanying risk of
audit and litigation), Organizations can transition to OSS/FS in part or in stages, which for many is a far
more practical transition approach.
Realizing these potential OSS/FS
benefits may require approaching problems in a different way. This might include
using thin clients, deploying a solution by adding a feature to an OSS/FS
product, and understanding the differences between the proprietary and OSS/FS
models. Acquisition processes may need to change to include specifically
identifying OSS/FS alternatives, since simply putting out a “request for
proposal” may not yield all the viable candidates. OSS/FS products are not the
best technical choice in absolutely all cases, of course; even organizations
which strongly prefer OSS/FS generally have some sort of waiver process for
proprietary programs. However, it’s clear that considering OSS/FS alternatives
can be beneficial.
Of course, before deploying any program you need to
evaluate how well it meets your needs, and some organizations do not know how to
evaluate OSS/FS programs. If this describes your circumstance, you may wish to
look at the companion articles How to Evaluate OSS/FS Programs
and the Generally Recognized
as Mature (GRAM) list.
OSS/FS options should be carefully
considered any time software or computer hardware is needed. Organizations
should ensure that their policies encourage, and not discourage, examining
OSS/FS approaches when they need software.
Microsoft's latest FUD is
its "Get the facts" campaign. Wheeler's definitive response, Why Open Source Software / Free
Software (OSS/FS), provides the facts, which show that Linux is equal to or
superior to Windows from the point of view of someone deciding whether to switch
from Windows to Linux, and OSS/FS software in general is equal to or superior to
competing proprietary software.
My previous post here also quotes his
conclusions.
[ Reply to This | # ]
|
|
Authored by: OldNerdGuy on Thursday, January 08 2004 @ 01:20 PM EST |
Read the Microsoft site listed. They figured out that Windows on an Intel Server
costs less than Linux on a Mainframe...
Gee whiz, and a used Yugo costs less than a Formula 1 Racecar.
DOOH!!![ Reply to This | # ]
|
|
Authored by: JMonroy on Thursday, January 08 2004 @ 01:39 PM EST |
I was just getting around to reading my January 2004
edition of Scientific
American. There is multi-age
advertisement from the Korean government
touting it's
industry and technology in-roads (which, incidentally, is
expanding by leaps and bounds).
Going to the section titled,
"SAMSUNG Unleashes the Future
of Technology," there is something that caught
my eye:
"[..] A complete 'System-On-Chip,' the mobile
application processor will reduce overall system costs and
eliminate the
need to configure additional components.
[..] It also supports major
operating systems
including Microsoft Windows CE, Palm OS, Symbian and
Linux."
Without reading too much into this
advertisement, I still
find it interesting that a major computer hardware
manufacturer even bothered to mention Linux. Sounds to me
like the writing
is on the wall for Microsoft... [ Reply to This | # ]
|
|
Authored by: Anonymous on Thursday, January 08 2004 @ 01:53 PM EST |
Interesting. When I follow PJ's "Get the Facts on Windows and Linux"
link
(http://www.microsoft.com/mscorp/facts/default.asp), I get a
page not found
message if I go after the page with netscape 4.7. With
IE, there is indeed a
page there. Is this some kind of attempt to
hide the page from those who might
not already have a Microsoft
bent? [ Reply to This | # ]
|
|
Authored by: Anonymous on Thursday, January 08 2004 @ 02:12 PM EST |
There is an extremly comprehensive paper on OSS, FSF, etc. at Free Software / Open Source: Information Society
Opportunities for Europe. This paper was published in 2000, and at one point
covers the use of API header files, result of use is generally ok.
It is a must
read. [ Reply to This | # ]
|
|
Authored by: Anonymous on Thursday, January 08 2004 @ 03:01 PM EST |
eweek opinion, about Microsoft's 'Get The Facts' campaign...
Get The Fud
I
hope it is not a duplicate post... [ Reply to This | # ]
|
|
Authored by: Anonymous on Thursday, January 08 2004 @ 03:23 PM EST |
"they paid $60,000 to BEA for development tool"
Another bad decision since the main tool that BEA sells is Enterprise Java Beans
stuff that you can get free (and better) from JBOSS.
See http://www.cioupdate.com/news/article.php/1581521
That begins:
February 10, 2003: Shares of BEA Systems briefly dropped nearly 10 percent last
week, as word spread on Wall Street of recent gains by the open source
application server Jboss.
As PJ says -- hard to compete with free.
JBoss business model is free software with support, training and documentation
available for good ole greenbacks.[ Reply to This | # ]
|
|
Authored by: Anonymous on Thursday, January 08 2004 @ 04:15 PM EST |
In previous articles Hey.Quic
k! Anybody go a Lid for this Coffin? and MS:Linu
x will be Hounded Over IP for 4 to 5 Years PJ covered the hidden cost of
using M$ SQL Server 7 product. This is the liability to patent infringement
from Timeline, the owners of the product.
M$ only licensed it, and did
not get all of the rights of use. In a court case Timeline prevailed and have
threatened to sue all users who have infringed on the patents. Needless to say,
M$ will indemnify all of the people that bought the SQL Server 7 product .
The first details of the latest ruling are from 02/20/2003.
These old
articles are from a time when Groklaw was not much travelled, and so there are
very few comments. Also, it appears that most of the current readers were not
reading then.
By the way, is the person in the coffin PJ? [ Reply to This | # ]
|
|
Authored by: Anonymous on Thursday, January 08 2004 @ 04:40 PM EST |
If IBM announces that they are shutting down all Windows development, or
"end-of-life" the effort and support for Windows, the game is over.
Seems fair, after Microsoft just a few days went to the great trouble to say
that IBM is ending "support" for OS/2. Isn't it nice when
competitors make your press annoucnements for you? Wouldn't it be even nicer
if they could get the facts straight?
IBM has more Windows app programmers than Microsoft - and if IBM management
pulls the plug on their work, Microsoft will not be able to stop the
hemoraging.
As for TCO: Microsoft is lying. NOBODY outside Microsoft "owns" a
copy of Windows or any of their proprietary apps. How can cost of OWNERSHIP be
less than or greater than x, when you simply cannot play in the game at all
because YOU can't own it?[ Reply to This | # ]
|
|
Authored by: phrostie on Thursday, January 08 2004 @ 04:42 PM EST |
for several years, there has been talk of Linux ports of catia(high end cad).
there was an article in LJ that mentioned that it existed, but IBM was holding
off on it's release. then back 08/2002 there was a post to one of the news
groups that had links to the IBM page:
http://www.research.ibm.com/deepview/index.html
complete with screenshots of various graphical applications(including catia)
running on nonless than RedHat Linux.
who would not want this for their desktop.
---
=====
phrostie
Oh I have slipped the surly bonds of DOS
and danced the skies on Linux silvered wings.
http://www.freelists.org/webpage/cad-linux[ Reply to This | # ]
|
|
Authored by: PeteS on Friday, January 09 2004 @ 05:56 AM EST |
There is an informative article Leave it to
beaver at The Guardian
Leave it
to beaver is a reference to Linus comment on the release of 2.6.
Interesting
reading, comments briefly on SCOX and it's apparent zero effect on those who are
still doing the coding.
--- Artificial Intelligence is no match for
natural stupidity [ Reply to This | # ]
|
|
|
|
|