SCO now turning on their own dwindling customer base...
FyRE
Well, it seems that SCO must be running out of people, companies and
countries to offend with their increasingly desparate attempts to pump
the stock price. Now they're threatening their own customers with
legal action if they dare to use Linux. So all you resellers out there
dancing between the two sides of "the force". It's time to make a
descision. Do you go over to the dark side with Darl Vader and his
bumbling court-troopers or do the right thing and drop SCO? After all,
if you're doing both then by SCO's definition, you're an accessory to
a crime.
Man, SCO must really have some pussy-whipped, browbeaten, spineless
wimps for customers if they actually put up with insults and threats
from their OS supplier.
For myself, I'd just love for the header files in linux to be updated
(ok, it would mean recompiling a lot of software, but hey, we have the
source code and it'd be worth it ;-) just enough to completely break
compatibility with SCO's products. I think it'd be hilarious to have
absolutely *zero* GNU/Open Source software running under SCO OS'.
--
FyRE < "War: The way Americans learn geography" >
J. L. Schilling
>
> For myself, I'd just love for the header files in linux to be updated
> (ok, it would mean recompiling a lot of software, but hey, we have the
> source code and it'd be worth it ;-) just enough to completely break
> compatibility with SCO's products. I think it'd be hilarious to have
> absolutely *zero* GNU/Open Source software running under SCO OS'.
You are betraying an ignorance of how portable software works.
Most GNU or other open source software is already coded to compile
and run under many different OSes. This is done in source code
via #ifdef's, configure scripts that auto detect the OS, and the like.
Changing characteristics of Linux will do *nothing* to change how the
software builds and runs on other OSes.
To remove SCO support, you'd have to change the parts of the code
that deals specifically with SCO OSes. This isn't easy because SCO
OSes are very similar to other Unix flavors, for example Solaris and
UnixWare are very close. If you rip out UnixWare support you could
easily rip out Solaris support at the same time.
So you'd really need to add explicit #ifdef's for SCO OSes and
sabotage the program in that code. But since you have to redistribute
these modifications under GPL or the other licenses, someone wanting to
build on SCO could easily find these sections and take them back out.
Hardly worth the effort.
Face it, one of the main things you give up under the open source model
is any control over how or where your software will be used.
Jonathan Schilling
Bill Burns
> http://sco.com/scosource/unix_licensee_letter_20031218.pdf
>
> Well, it seems that SCO must be running out of people, companies
> and countries to offend with their increasingly desparate attempts
> to pump the stock price. Now they're threatening their own
> customers with legal action if they dare to use Linux.
Is it possible that *any* Unix licensee did not throw this letter
immediately in the garbage?
--
Bill Burns, Long Island, NY, USA
mailto:bi...@ftldesign.com
History of Technology Websites:
http://ftldesign.com
Allan Chandler
"FyRE" <Fy...@toktik.demon.ku.oc.x> wrote in message
news:2nklvvgftkfn0rnv1...@4ax.com...
>
> Well, it seems that SCO must be running out of people, companies and
> countries to offend with their increasingly desparate attempts to pump
> the stock price. Now they're threatening their own customers with
> legal action if they dare to use Linux.
I think those letters went to SCO's _source code_ licencees, not people
using SCO products.
Brian
>> Well, it seems that SCO must be running out of people, companies
>> and countries to offend with their increasingly desperate attempts
>> to pump the stock price. Now they're threatening their own
>> customers with legal action if they dare to use Linux.
> Is it possible that *any* Unix licensee did not throw this letter
> immediately in the garbage?
These letters were sent to THEIR customers, companies that are licensed to
use SCO IP. It is a requirement of that license agreement that SCO has the
right to audit for compliance. I am guessing that demanding confirmation of
the terms may be within their rights however I doubt SCO will be around
long enough to actually harass all their licensees that fail to make a
satisfactory declaration.
It is hard to imagine a better way to alienate your clients!
8^)
We should have a newsgroup feature called: Stupid SCO Tricks!
Anybody want to speculate on the next stupid SCO trick?
I think SCO has until Monday to produce the documents required by IBM's
successful Motion To Compel - Can we expect a flurry of IBM Motions To
Strike?
8^)
One last thought, during the aforementioned hearing on the Motion To Compel,
McBride stated that new copyright claims would be filed within a week or
two - them weeks has passed and no new claims have been filed. McBride lied
to a judge - not a smart thing to do, IMHO.
Happy New Year - almost certainly the last year of SCO's existence.
Brian
Brian K. White
Take another bong hit and next tell me the one about how the govermint
is like paying the RIAA to be the front for their plans to eventually
phase out the concept of privacy and personal liberty and stuff.
FyRE
The thought hadn't crossed my mind, "Brain". I'm just reporting the
facts here. Now run along...
Joe Dunning
>Bill Burns wrote:
>> FyRE wrote:
>>> http://sco.com/scosource/unix_licensee_letter_20031218.pdf
>
>>> Well, it seems that SCO must be running out of people, companies
>>> and countries to offend with their increasingly desperate attempts
>>> to pump the stock price. Now they're threatening their own
>>> customers with legal action if they dare to use Linux.
>
>> Is it possible that *any* Unix licensee did not throw this letter
>> immediately in the garbage?
>
>These letters were sent to THEIR customers, companies that are licensed to
>use SCO IP. It is a requirement of that license agreement that SCO has the
>right to audit for compliance. I am guessing that demanding confirmation of
>the terms may be within their rights however I doubt SCO will be around
>long enough to actually harass all their licensees that fail to make a
>satisfactory declaration.
>
I think that there is a right to audit source code licenses, the
question I have is: is that SCO's right or Novell's?
Also, it's pretty clear that the audit rights don't include all the
demands that SCO is making. I would think that the most obvious answer
would be to refer SCO back to the license agreement and answer the
specific audit requirements from the agreement.
>One last thought, during the aforementioned hearing on the Motion To Compel,
>McBride stated that new copyright claims would be filed within a week or
>two - them weeks has passed and no new claims have been filed. McBride lied
>to a judge - not a smart thing to do, IMHO.
According to the transcripts that I have seen, McBride stated that they
would add copyright claims in *more* than a week. Probably when hell
freezes over, I guess.
Brian
> Take another bong hit and next tell me the one about how the govermint
> is like paying the RIAA to be the front for their plans to eventually
> phase out the concept of privacy and personal liberty and stuff.
That's Tony's theory - check out his website for some truly excellent
conspiracy theories.
And I don't believe Tony even has a bong.
8^)
Brian
Brian
>>These letters were sent to THEIR customers, companies that are licensed to
>>use SCO IP. It is a requirement of that license agreement that SCO has the
>>right to audit for compliance. I am guessing that demanding confirmation
>>of the terms may be within their rights however I doubt SCO will be around
>>long enough to actually harass all their licensees that fail to make a
>>satisfactory declaration.
> I think that there is a right to audit source code licenses, the
> question I have is: is that SCO's right or Novell's?
Good question...
I would guess that would be part of the rights conveyed.
> Also, it's pretty clear that the audit rights don't include all the
> demands that SCO is making. I would think that the most obvious answer
> would be to refer SCO back to the license agreement and answer the
> specific audit requirements from the agreement.
That's SCO for ya...
>>One last thought, during the aforementioned hearing on the Motion To
>>Compel, McBride stated that new copyright claims would be filed within a
>>week or two - them weeks has passed and no new claims have been filed.
>>McBride lied to a judge - not a smart thing to do, IMHO.
> According to the transcripts that I have seen, McBride stated that they
> would add copyright claims in *more* than a week. Probably when hell
> freezes over, I guess.
Here is an exact quote from the transcript at Groklaw:
www.groklaw.net/article.php?story=2003121122033016
McBride: ... I will proffer to the Court that we are filing a second amended
complaint that has copyright infringement claims, and will be filed within
the coming few days or no less than a week.
Wow - that's a lie.
8^)
Best regards,
Brian
Joe Dunning
>
>To remove SCO support, you'd have to change the parts of the code
>that deals specifically with SCO OSes. This isn't easy because SCO
>OSes are very similar to other Unix flavors, for example Solaris and
>UnixWare are very close. If you rip out UnixWare support you could
>easily rip out Solaris support at the same time.
I would not be surprised if quite a few FOSS projects are removing any
explicit support for UnixWare and OpenServer right now. In other words,
rather than trying to deliberately break them for SCO OS-es, they may be
trying to remove any code that is *only* required for SCO OS-es.
As you point out, this does not stop anyone putting such support back
in, but it does impose more of a burden.
Disclaimer: the above is 100% speculation!
FyRE
wrote:
[...]
>McBride: ... I will proffer to the Court that we are filing a second amended
>complaint that has copyright infringement claims, and will be filed within
>the coming few days or no less than a week.
>
>Wow - that's a lie.
Try re-reading: "or no less than a week".
Brian
> On Wed, 07 Jan 2004 08:31:54 GMT, Brian <br...@stanley-park.com>
> wrote:
> [...]
>>McBride: ... I will proffer to the Court that we are filing a second
>>amended complaint that has copyright infringement claims, and will be
>>filed within the coming few days or no less than a week.
>>Wow - that's a lie.
> Try re-reading: "or no less than a week".
Ok - a funny play on words.
It has been over 3 weeks but I guess it's kind of open ended.
So, what do you think the 23rd will bring?
8^)
Brian
J. L. Schilling
>
> I would not be surprised if quite a few FOSS projects are removing any
> explicit support for UnixWare and OpenServer right now. In other words,
> rather than trying to deliberately break them for SCO OS-es, they may be
> trying to remove any code that is *only* required for SCO OS-es.
>
> As you point out, this does not stop anyone putting such support back
> in, but it does impose more of a burden.
Trouble is, most open source projects also have open cvs trees as well.
So all someone has to do is reverse one cvs change and presto the SCO
support is back in. So to counteract that you'd to obfuscate the change
or the comments for the change, to hide the removal of support ...
but of course this makes the program less understandable and maintainable,
and pretty soon some other developer trips over the obfuscation and introduces
a bug that affects users on all platforms. So all you've done is made
your own life more miserable.
Jonathan Schilling
FyRE
Schilling) wrote:
Well you're presuming that the CVS commit would be solely to break SCO
support. But what if that were part of a much bigger change? Also,
since this would then become part of the CVS source tree, you'd then
have to maintain your own fork for subsequent patches.
Personally I'd advocate removing all the ifdefs for SCO, and pull in a
bunch of Linux/*BSD kernel header files for each project along with
the latest libraries. That way, anyone running a Linux or BSD kernel
would have no problems building the project (so long as they had the
kernel source) and the latest GNU libs.
Actually I'd go further and suggest the most important projects
specifically changed their licences to exclude the SCO platform.
Certain parts of the projects would have to remain GPL'ed, but not
all.
Joe Dunning
>j...@blahblah.invalid (Joe Dunning) wrote in message news:<oyPKb.148966$VB2.577362@attbi_s51>...
>>
>> I would not be surprised if quite a few FOSS projects are removing any
>> explicit support for UnixWare and OpenServer right now. In other words,
>> rather than trying to deliberately break them for SCO OS-es, they may be
>> trying to remove any code that is *only* required for SCO OS-es.
>>
>> As you point out, this does not stop anyone putting such support back
>> in, but it does impose more of a burden.
>
>Trouble is, most open source projects also have open cvs trees as well.
You are claiming that anyone can commit changes to most open source
projects? I don't think so. Many or most have anonymous READ access, but
commiting changes is limited to the project developers.
Joe Dunning
The question might be: what will the 10th bring? Since SCO is required
to provide answers and specific allegations by then. It's just that the
next hearing is on the 23rd. SCO already asked for an extension beyond
the 10th and it was denied.
Bill Campbell
...
>The question might be: what will the 10th bring? Since SCO is required
>to provide answers and specific allegations by then. It's just that the
>next hearing is on the 23rd. SCO already asked for an extension beyond
>the 10th and it was denied.
No surprise there. If I remember correctly the Judge was very explicit
that the 30 days was a hard number, not subject to delay.
Bill
--
INTERNET: bi...@Celestial.COM Bill Campbell; Celestial Software LLC
UUCP: camco!bill PO Box 820; 6641 E. Mercer Way
FAX: (206) 232-9186 Mercer Island, WA 98040-0820; (206) 236-1676
URL: http://www.celestial.com/
The only logical reason to take guns away from responsible people is to
give irresponsible people an edge in the perpetration of their crimes
against us. -- The Idaho Observer, Vol. 1, No. 2 February 1997
J. L. Schilling
>
> Personally I'd advocate removing all the ifdefs for SCO, and pull in a
> bunch of Linux/*BSD kernel header files for each project along with
> the latest libraries. That way, anyone running a Linux or BSD kernel
> would have no problems building the project (so long as they had the
> kernel source) and the latest GNU libs.
First of all, most normal applications have no need for kernel
header files. Kernel header files are for building the kernel!
User-space applications pull in header files that describe the
OS APIs (such as read(), pthread_create(), networking calls,
X11 calls, etc etc). [A few header files typically are shared
between the kernel and the libraries, such as those that
describe fundamental types, but in general this distinction holds.]
So when you're porting an application to an OS, you generally
don't care what the kernel looks like; you care about the user-space
stuff, what's in /usr/include/ and /usr/lib/ and whatnot. That's
good, because it means the kernel can undergo wholesale revisions
without disrupting application code (and the Linux world does take
advantage of this freedom, as do all OS makers). And it means that if
your application conforms to the relevant POSIX and Single UNIX
Specification standards, you can port it from OS A to OS B with
little or no changes.
Now, if your plan is to force a dependency between applications
and the underlying kernel they will run on, such that you can't
even build the app without having pieces of the kernel available
in the app's build tree ... well, that's going to make life
totally difficult, even for people who are just interested in moving
from one implementation of Linux to another. Indeed your plan would
damage the open source software world far more than anything SCO
might ever do!
Similarly, your idea to pull in specific library binaries into
an application's distribution is usually a bad idea. This is
equivalent to linking binaries statically rather than dynamically.
But what assurance do you have that these binaries will function
correctly on a different revision of the same OS? Better is to
take advantage of dynamic libraries. Then your link to, say,
libsocket.so will resolve to whatever libsocket is currently
installed on that system, which is presumably the right one
to interface to that kernel's networking primitives. (To be sure,
there used to be a school of thought that static linking was
better because it allowed you to test the system library that you
knew would be used. But experience has generally shown that
dynamic linking is better, which is why you see linkers defaulting
to that, and why .so's of system libraries are used more than .a's.)
Jonathan Schilling
FyRE
Schilling) wrote:
>FyRE <Fy...@toktik.demon.ku.oc.x> wrote in message news:<cljqvv4pqe062qd6r...@4ax.com>...
>>
>> Personally I'd advocate removing all the ifdefs for SCO, and pull in a
>> bunch of Linux/*BSD kernel header files for each project along with
>> the latest libraries. That way, anyone running a Linux or BSD kernel
>> would have no problems building the project (so long as they had the
>> kernel source) and the latest GNU libs.
>
>First of all, most normal applications have no need for kernel
>header files. Kernel header files are for building the kernel!
>User-space applications pull in header files that describe the
>OS APIs (such as read(), pthread_create(), networking calls,
>X11 calls, etc etc). [A few header files typically are shared
>between the kernel and the libraries, such as those that
>describe fundamental types, but in general this distinction holds.]
>So when you're porting an application to an OS, you generally
>don't care what the kernel looks like; you care about the user-space
>stuff, what's in /usr/include/ and /usr/lib/ and whatnot. That's
>good, because it means the kernel can undergo wholesale revisions
>without disrupting application code (and the Linux world does take
>advantage of this freedom, as do all OS makers). And it means that if
>your application conforms to the relevant POSIX and Single UNIX
>Specification standards, you can port it from OS A to OS B with
>little or no changes.
Erm, yes I'm aware of all this. The idea is break compatibility with
binary-only kernels (ie, SCO). I've written a good bit of C in my
time, and modified the source of some Linux tools for myself...
>Now, if your plan is to force a dependency between applications
>and the underlying kernel they will run on, such that you can't
>even build the app without having pieces of the kernel available
>in the app's build tree ... well, that's going to make life
>totally difficult, even for people who are just interested in moving
>from one implementation of Linux to another. Indeed your plan would
>damage the open source software world far more than anything SCO
>might ever do!
No, Linux IS the kernel. Thus, the headers will be exactly the same
across distros. This is assuming one doesn't pull in headers from
patched modules etc. A lot of headers haven't changed in years, BTW...
>Similarly, your idea to pull in specific library binaries into
>an application's distribution is usually a bad idea. This is
>equivalent to linking binaries statically rather than dynamically.
I was refering to linking to the latest dynamic shared libraries.
Since SCO are usually a good way behind Linux, it would make life a
little bit more difficult to port stuff across..
>But what assurance do you have that these binaries will function
>correctly on a different revision of the same OS? Better is to
>take advantage of dynamic libraries. Then your link to, say,
>libsocket.so will resolve to whatever libsocket is currently
>installed on that system, which is presumably the right one
>to interface to that kernel's networking primitives. (To be sure,
>there used to be a school of thought that static linking was
>better because it allowed you to test the system library that you
>knew would be used. But experience has generally shown that
>dynamic linking is better, which is why you see linkers defaulting
>to that, and why .so's of system libraries are used more than .a's.)
The reason it's thought of as better is due to memory usage. Building
statically does actually have an advantage for applications that you'd
only tend to have a single instance of running at a time. The startup
time is often drastically reduced for one thing. Another option is
prelinking: Some benchmarks when doing this with my favourite distro
are here: http://www.gentoo.org/main/en/performance.xml (Mozilla loads
twice as quickly, and Kmail 10 times as fast).
J. L. Schilling
> Erm, yes I'm aware of all this. The idea is break compatibility with
> binary-only kernels (ie, SCO). I've written a good bit of C in my
> time, and modified the source of some Linux tools for myself...
My browser crashed just as I was posting a long reply to this,
so I don't know if it got out. Grrr.
The main points:
-- Sorry if my previous tone was patronizing.
-- Still don't understand in detail what your scheme is to break
open source apps on non-open kernels. Use a C hello world program
as a simple example of a open source app. How would you scheme work?
-- Agree that if app developers use library *APIs* that exist only on
Linux, that will make porting to other OSes harder. But that has
nothing to do with library binaries or linking style.
-- Dynamic linking may be a loss for any one application measured,
but for the kind of big multiapp server systems that UnixWare 7 runs
on, it's a big win in overall throughput. Reason is that only one
copy of libc has to be allocated in memory, that all running programs
share. Ditto libm, libthread, libsocket, X libs, JVM libs, etc.
Each app sacrifices a little on its own, overall throughput benefits
greatly.
-- Thanks for the pointer on "prelinking". Will look at it some more.
Jonathan Schilling
Joe Dunning
wrote:
>FyRE <Fy...@toktik.demon.ku.oc.x> wrote in message news:<qdltvvo87r8hhule5...@4ax.com>...
>> Erm, yes I'm aware of all this. The idea is break compatibility with
>> binary-only kernels (ie, SCO). I've written a good bit of C in my
>> time, and modified the source of some Linux tools for myself...
>
>-- Agree that if app developers use library *APIs* that exist only on
> Linux, that will make porting to other OSes harder. But that has
> nothing to do with library binaries or linking style.
I don't think anyone would argue that Open Source programs can be
written so that it is impossible to port them to any Unix-like OS.
My original suggestion was that the FOSS developers could impose an
extra burden to port to UnixWasre and OpenServer. Not that it could be
made impossible.
Bill Campbell
And how does this hurt SCO? They're doing a great job of committing
corporate suicide on their own, and I doubt that the bulk of their
OpenServer business depends on open source or things like Samba (how useful
is this on a cash register).
Making open source more difficult to build on SCO Unix probably hurts their
installed base of SMB customers much more than it does SCO, and many of
these customers are at the mercy of 3rd party software developers.
Bill
--
INTERNET: bi...@Celestial.COM Bill Campbell; Celestial Software LLC
UUCP: camco!bill PO Box 820; 6641 E. Mercer Way
FAX: (206) 232-9186 Mercer Island, WA 98040-0820; (206) 236-1676
"I do not feel obliged to believe that the same God who has endowed us
with sense, reason, and intellect has intended us to forego their use."
-- Galileo Galilei
Joe Dunning
>On Sat, Jan 10, 2004, Joe Dunning wrote:
>>On 9 Jan 2004 20:43:03 -0800, J. L. Schilling <jlsels...@my-deja.com>
>>wrote:
>>
>>
>>I don't think anyone would argue that Open Source programs can be
>>written so that it is impossible to port them to any Unix-like OS.
>>My original suggestion was that the FOSS developers could impose an
>>extra burden to port to UnixWasre and OpenServer. Not that it could be
>>made impossible.
>
>And how does this hurt SCO? They're doing a great job of committing
>corporate suicide on their own, and I doubt that the bulk of their
>OpenServer business depends on open source or things like Samba (how useful
>is this on a cash register).
>
>Making open source more difficult to build on SCO Unix probably hurts their
>installed base of SMB customers much more than it does SCO, and many of
>these customers are at the mercy of 3rd party software developers.
I think your arguments are possibly contradictory. As has been pointed
out, changes to FOSS code will not make the old versions magically
disappear or cease working. Thus existing installations are unlikely to
be affected. From what I have ready, many installations are not
accessible from the Internet and remote intrusions are rarely considered
a threat (I stress: "many installations" -- not all). Hence the lack of
updates is not going to be a threat to many installations.
On the other hand, if there is an impact on customers, those customers
will pressure the 3rd party developers to support other platforms. I've
seen a lot of postings that imply that 3rd party support is frequently
the only reason to use SCO's products.
Finally, if support of SAMBA and other FOSS tools is not important, why
was the announcement of support of Samba 3.0 given such prominance at
SCOForum?
Actually, I was thinking that a more subtle approach might not be to
remove support, but make changes that impact performance only on SCO
systems. Probably not a very practical idea, though.
As for SCO committing suicide -- while this is true, I don't see why
SCO should be allowed to continue the anti-Linux FUD any longer.
Brian K. White
> On Sat, Jan 10, 2004, Joe Dunning wrote:
> >On 9 Jan 2004 20:43:03 -0800, J. L. Schilling <jlsels...@my-deja.com>
> >wrote:
> >
> >>FyRE <Fy...@toktik.demon.ku.oc.x> wrote in message news:<qdltvvo87r8hhule5...@4ax.com>...
> >>> Erm, yes I'm aware of all this. The idea is break compatibility with
> >>> binary-only kernels (ie, SCO). I've written a good bit of C in my
> >>> time, and modified the source of some Linux tools for myself...
> >>
> >>-- Agree that if app developers use library *APIs* that exist only on
> >> Linux, that will make porting to other OSes harder. But that has
> >> nothing to do with library binaries or linking style.
> >
> >I don't think anyone would argue that Open Source programs can be
> >written so that it is impossible to port them to any Unix-like OS.
> >My original suggestion was that the FOSS developers could impose an
> >extra burden to port to UnixWasre and OpenServer. Not that it could be
> >made impossible.
>
> And how does this hurt SCO? They're doing a great job of committing
> corporate suicide on their own, and I doubt that the bulk of their
> OpenServer business depends on open source or things like Samba (how useful
> is this on a cash register).
Well, our installations are all SCO, and all use file & print sharing
a lot, but we use FacetWin for that for the same reason we use SCO for
the OS.
(reliability and stability, not resale margin, for the "commercial
software is evil" contingent. we only charge a token margin on
hardware & re-sold software.)
FyRE
Stupidity?
Brian K. White
That's it! Exactly! Oh my GOD I've been so stupid all these years!
What was I thinking? I never even LOOKED at any other platforms
before. SCO told me there is no such thing and I beleived them like
the stupid sheep that I am. Thank you, Oh Thank You for opening up
mine eyes to the one true light! How would the world get along without
FyRE? I mean, since the current management of sco are mental and
ethical lepers, that OBVIOUSLY proves that for the last 15 years or so
all us vars have actually been hallucinating all that dependable
service! And all thos old Xenix boxes that are even now still clanking
away since long before linux was so much as a wet dream must actually
also be some kind of hoax perpetrated for the sole purpose of duping
stupid people for some reason that I'm too stupid to fathom myself
since I was one of them. Hmmm, come to think of it.. has anyone ever
seen Darl McBride and Bill Gates in the same room at the same time???
very interesting... Wow, come to think of it, Maybe it's no
coincidence that Darl is in the media a lot lately, and they can't
find Osama Bin Laden anywhere! Clearly Darl and Bill and Osama are all
the same person!
I must take an axe to all my sco boxes immediately before judgement
day comes!
Thank you again oh wisest of the wise.
Scott Burns
>I don't think anyone would argue that Open Source programs can be
>written so that it is impossible to port them to any Unix-like OS.
>My original suggestion was that the FOSS developers could impose an
>extra burden to port to UnixWasre and OpenServer. Not that it could be
>made impossible.
>
>
>
Most people here seem to be of the opinion that SCO do not have a leg to
stand on. Most people here seem to think that unless SCO have something
major up their sleeves they are going to lose. It seems to me that most
debate here centers on the likelyhood of SCO having surprises in store.
If SCO do win, I put it to you that compatibility is the least of our
worries. If they lose, how many hours of work is essentially wasted?
It might feel good at the time, help people feel they are fighting
back, but in a years time, win or lose, what will they have accomplished?
I've been involved with between 75 and 100 SCO systems over the years.
Not one was kept up to date with patches. Less than 10 have FOSS
compiled specifically for them as opposed to skunkware. I have no doubt
that some people will still be running SCO in 10 years time even if
OpenServer was removed from sale today. In the great big scheme of
things, just what portion of systems will be affected by making it
harder to compile modern versions of programs?
Just how much effort is really justified here?
--
Scott Burns
Mirrabooka Systems
Tel +61 7 3857 7899
Fax +61 7 3857 1368
J. L. Schilling
>
> [...] Another option is
> prelinking: Some benchmarks when doing this with my favourite distro
> are here: http://www.gentoo.org/main/en/performance.xml (Mozilla loads
> twice as quickly, and Kmail 10 times as fast).
Interesting. The results seem uneven when applied
(see http://www.crasta.com/james/articles/prelink.php),
but that's not unusual with an optimization of this sort.
Lazy binding of dynamic function references as done on UW7
(and Linux) will reduce some of the need for this. Also,
many modern apps have a dlopen()-based DSO architecture (Apache
and Java virtual machines are two common examples), where
resolving doesn't have to be done unless the .so is actually
loaded (which often never happens). The name conflict
resolution in this dlopen case must get a bit messy in the
prelinking scheme (as described in the Cygnus paper
at http://people.redhat.com/drepper/lt2002talk.pdf).
Also, Linux tends to have more desktop users than UW7,
and thus more need for fast startup of tool apps.
Nevertheless this is important on UW7 too, and various
optimizations towards this have been done over the years,
from executable/library juxtaposition via ld options to X11
library conglomeration to the "fur" binary editing optimization tool
to JIT selective compilation heuristics in the Java virtual machine
(http://www.sco.com/developers/java/news/jit-heur.pdf).
Jonathan Schilling
Skot
>
<sweep>
> --
> FyRE < "War: The way Americans learn geography" >
Wow! You guys really fall for this troll bait, don't you?
LMAO
Brian K. White
Not really.
Like Tony said in another thread when someone else aksed roughly the
same question:
It's not that we care the slightest bit what these few inconsequencial
wackos think. It's that there are a lot of quiet readers who do not
have as solid an understanding of the issues and the resulting solid
stance with regard to them, as those of us who actually have jobs that
require supporting a lot of customers running sco and/or other unices.
These quiet readers may get the impression that these wacko's
statements actually have some validity if no one seems to offer any
counter-argument, or simply point out the flaws in their arguments.
Someone should at least make it understood, so that it is not left to
be inferred wrongly, whether or not those of us who actually do know
what we are talking about and have opinions based on actual real-world
jobs supporting production boxes and giving people good service (where
"good service" means they are not susceptible to all the linux
exploits that keep coming around almost as bad as windows lately, and
you don't have to touch their boxes for months or years at a time)
consider their statements at all valid. These readers probably don't
have solid first-hand knowledge that would put the wackos statements
into perspective. If someone says something, and it sounds even
slightly plausible, and no one says anything else to dis-prove it,
then in many cases the reader must quite reasonably assume that the
original statement had some amount of value and truth. They don't
already know better on their own. So someone who _does_ know their
stuff should help people from being snowballed by helping them see the
whole picture and allow them to form an educated opinion that has
value instead of a default opinion that has no value or actually a
negative value.
Silence does NOT imply agreement or aquiescence, but the fact is, a
lot of people are not mentally logical enough to keep that in mind at
the emotional level. An unfortunately largish subset of people will
and do take silence to mean:
silence, thus not_disagree, thus agree.
Which is of course completely false.
And frankly, I don't need any more people in the world forming
backwards and counter-productive attitudes based on the bad info some
people fling around without caring what damage they are doing or who
they are hurting. (They hurt end-users by getting them to think they
have to do things and put up with things they don't actually have to.)
Every time some idiot convinces a customer they should be running
their main
application on Windows, it makes my life harder because I then have to
not only educate them about all the obvious and subtle and direct and
indirect ways in which relying on unix is more sane than relying on
windows, I also have to de-educate them about the false promises they
were already given and convinced of.
Generally, people don't _really_ "get it" until their windows server
farts and bletcherates their data or costs them several thousand bucks
in anti-virus-related labor and down-time that would have never
happened on unix. I can't stand it when that happens. It's a waste of
time, money, effort, and brains, that could really just about be
considered criminal.
(What else do you call it when you make a promise, take the money, and
then don't deliver, and charge more money to fix problems that were
promised not to happen in the first place?) To a lesser extent, the
over-zealous militant linux contingent suffer from the same false
foundation. Linux is indeed a better foundation than windows, but if
you try to claim it's an equivalent replacement for a good solid
commercial unix, you are either lying, or at best merely ignorant of
the larger real world where stability matters above all else.
(stability over time and across versions, not just stability as in
long uptimes, although of course that is important too)
So we offer counter-statements pretty much merely to keep the wackos
voices from being the only voices and to ensure that at the very least
the counter-proofs are on record.
Joe Dunning
>Skot <sk...@canada.com> wrote in message news:<4005715E...@canada.com>...
>> FyRE wrote:
>> >
>> <sweep>
>>
>> > --
>> > FyRE < "War: The way Americans learn geography" >
>>
>> Wow! You guys really fall for this troll bait, don't you?
>>
>> LMAO
>
>
>Not really.
>Like Tony said in another thread when someone else aksed roughly the
>same question:
>
>It's not that we care the slightest bit what these few inconsequencial
>wackos think. It's that there are a lot of quiet readers who do not
>have as solid an understanding of the issues and the resulting solid
>stance with regard to them, as those of us who actually have jobs that
>require supporting a lot of customers running sco and/or other unices.
Did it ever occur to you that other people might view things the other
way round? When I started reading this newsgroup, there was no mention
of the lawsuit and the risks SCO is facing. Did it not occur to you that
some people might want to make sure that the quiet readers did not think
that all was well in SCO-land?
>To a lesser extent, the
>over-zealous militant linux contingent suffer from the same false
>foundation. Linux is indeed a better foundation than windows, but if
>you try to claim it's an equivalent replacement for a good solid
>commercial unix, you are either lying, or at best merely ignorant of
>the larger real world where stability matters above all else.
>(stability over time and across versions, not just stability as in
>long uptimes, although of course that is important too)
Clearly you have to exclude uptimes, since, if you look for the longest
uptimes on Netcraft, you will see ONLY various BSDs and a couple of
Linux machines. Not a single instance of a commercial OS.
Stability over time and versions: yes it may be available, but don't
forget the price: slower development. I think one of the reasons is that
Linux has advanced quickly is that the developers are not afraid to
throw out old code and start again when it is obvious that the new
approach would pay large benefits. Look at the firewall approaches:
IPFWADM, IPCHAINS and now Netfilter/IPTABLES. Even so, there is still
limited support for IPFWADM in the 2.4 series kernels.
Bill Vermillion
Joe Dunning <joedunning12...@hotmail.com> wrote:
>On 16 Jan 2004 15:01:26 -0800, Brian K. White <br...@aljex.com> wrote:
>Clearly you have to exclude uptimes, since, if you look for the
>longest uptimes on Netcraft, you will see ONLY various BSDs
>and a couple of Linux machines. Not a single instance of a
>commercial OS.
Not exactly true. The top-20 sites have a preponderance of
BSD/OS - and THAT is a commercial version based on the original
4.4Lite. Only recently has it been dropped by WindRiver - so as a
just a couple of months ago there is no commercial BSD/OS left.
What is interesting is in the top-20 uptimes there are 18 BSD/OS
and two FreeBSD sites. And in the top 30 it waw 27 BSD/OS and
three FreeBSD.
Only 1 Linux system there coming in at 31. The BSD/OS was
reasonably priced the last time I saw prices on it - and the only
prices I remember from the past was their original $500 in binary
and $1000 with source. But they found their sales increased
drastically when the repriced it at almost exactly 1/2 of that of
the SCO product line.
They are execptionally stable machines and my servers typically go
over a year before I upgrade to a newer OS. Only one time in the
past two years was there a patch that required a kernel change and
a reboot. Other than that it was patch a module, and restart the
module with the OS continuing on. Longest I went without reboot
for upgraded was about 750 days.
--
Bill Vermillion - bv @ wjv . com
Bill Campbell
>On 16 Jan 2004 15:01:26 -0800, Brian K. White <br...@aljex.com> wrote:
>>Skot <sk...@canada.com> wrote in message news:<4005715E...@canada.com>...
>>> FyRE wrote:
>>> >
>>> <sweep>
>>>
>>> > --
>>> > FyRE < "War: The way Americans learn geography" >
>>>
>>> Wow! You guys really fall for this troll bait, don't you?
>>>
>>> LMAO
>>
>>
>>Not really.
>>Like Tony said in another thread when someone else aksed roughly the
>>same question:
>>
>>It's not that we care the slightest bit what these few inconsequencial
>>wackos think. It's that there are a lot of quiet readers who do not
>>have as solid an understanding of the issues and the resulting solid
>>stance with regard to them, as those of us who actually have jobs that
>>require supporting a lot of customers running sco and/or other unices.
My job has involved supporting a wide variety if *ix systems since 1982,
SCO systems since 1987, Linux since 1994 or so, FreeBSD extensively in the
last year or so, and a smattering of other *ix systems throughout.
The first customer I had when I founded Celestial in 1984 was then running
a customized version of Tandy's accounting software (mostly RealWorld) and
FilePro on a Radio Shack Model 16/6000, and is now running essentially the
same software on OpenServer 5.0.5. There's no good reason for them to
switch from SCO since their software works, and will probably continue to
do so until the business owners retire at which time the business will
probably close. Most of our SCO customers have moved to Linux or FreeBSD
since we started installing Linux in mission critical applications in
September 1997 ranging from small law offices to ISPs with 10s of thousands
of customers.
None of these customers running Linux have experienced the kinds of
instability or security problems Brian describes. Of course we have never
installed vanilla systems from any vendor, and I have found security
problems with the commercial versions of Unix that were far more dangerous
and harder to fix than anything I've found in Linux. When security
problems have been found on Linux systems, fixes have been available very
quickly and we can get all the systems we support updated within a day at
most (bind, openssh, openssl, and a few others have been the most frequent
culprits).
>Did it ever occur to you that other people might view things the other
>way round? When I started reading this newsgroup, there was no mention
>of the lawsuit and the risks SCO is facing. Did it not occur to you that
>some people might want to make sure that the quiet readers did not think
>that all was well in SCO-land?
>
>>To a lesser extent, the
>>over-zealous militant linux contingent suffer from the same false
>>foundation. Linux is indeed a better foundation than windows, but if
>>you try to claim it's an equivalent replacement for a good solid
>>commercial unix, you are either lying, or at best merely ignorant of
>>the larger real world where stability matters above all else.
>>(stability over time and across versions, not just stability as in
>>long uptimes, although of course that is important too)
>
>Clearly you have to exclude uptimes, since, if you look for the longest
>uptimes on Netcraft, you will see ONLY various BSDs and a couple of
>Linux machines. Not a single instance of a commercial OS.
>
>Stability over time and versions: yes it may be available, but don't
>forget the price: slower development. I think one of the reasons is that
>Linux has advanced quickly is that the developers are not afraid to
>throw out old code and start again when it is obvious that the new
>approach would pay large benefits. Look at the firewall approaches:
>IPFWADM, IPCHAINS and now Netfilter/IPTABLES. Even so, there is still
>limited support for IPFWADM in the 2.4 series kernels.
My main gripe with Linux over the years is that there seems to be little
care about backwards compatibility, and things tend to change gratuitously.
It's funny how things come up on the 'Net that seem to hit just in time --
there was a posting on a FreeBSD mailing list just this morning pointing to
a very interesting article on this very subject:
http://www.over-yonder.net/~fullermd/rants/bsd4linux/bsd4linux1.php
A couple of good quotes from this article:
`` BSD is what you get when a bunch of Unix hackers sit down
to try to port a Unix system to the PC. Linux is what you
get when a bunch of PC hackers sit down and try to write a
Unix system for the PC.''
``BSD is designed. Linux is grown.''
Bill
--
INTERNET: bi...@Celestial.COM Bill Campbell; Celestial Software LLC
UUCP: camco!bill PO Box 820; 6641 E. Mercer Way
FAX: (206) 232-9186 Mercer Island, WA 98040-0820; (206) 236-1676
URL: http://www.celestial.com/
``People who relieve others of their money with guns are called robbers. It
does not alter the immorality of the act when the income transfer is
carried out by government.''
Bill Campbell
>In article <KQ_Nb.81841$xy6.140563@attbi_s02>,
>Joe Dunning <joedunning12...@hotmail.com> wrote:
>>On 16 Jan 2004 15:01:26 -0800, Brian K. White <br...@aljex.com> wrote:
>
>>Clearly you have to exclude uptimes, since, if you look for the
>>longest uptimes on Netcraft, you will see ONLY various BSDs
>>and a couple of Linux machines. Not a single instance of a
>>commercial OS.
>
>Not exactly true. The top-20 sites have a preponderance of
>BSD/OS - and THAT is a commercial version based on the original
>4.4Lite. Only recently has it been dropped by WindRiver - so as a
>just a couple of months ago there is no commercial BSD/OS left.
>
>They are execptionally stable machines and my servers typically go
>over a year before I upgrade to a newer OS. Only one time in the
>past two years was there a patch that required a kernel change and
>a reboot. Other than that it was patch a module, and restart the
>module with the OS continuing on. Longest I went without reboot
>for upgraded was about 750 days.
The limiting factors for most of the machines we support are power outages
or equipment moves (all of our systems here have been up 91 days which was
the last power outage before I bought a generator that got us through the
last outage). One of our customers has a Linux box at the local Sprint CO
along with their Annex 8000 RACs where the Linux box is used primarily for
radius authentication, secondary MX forwarding, and DNS. This box had been
up for about 450 days until Sprint decided to reorganize their rack and
cables 203 days ago (Caldera eDesktop 2.4).
Bill
--
INTERNET: bi...@Celestial.COM Bill Campbell; Celestial Software LLC
UUCP: camco!bill PO Box 820; 6641 E. Mercer Way
FAX: (206) 232-9186 Mercer Island, WA 98040-0820; (206) 236-1676
URL: http://www.celestial.com/
``Never do your enemy a minor injury.''
- Machiavelli
Tom Podnar
>
> What is interesting is in the top-20 uptimes there are 18 BSD/OS
> and two FreeBSD sites. And in the top 30 it waw 27 BSD/OS and
> three FreeBSD.
>
I wonder how many of the Netcraft listed "FreeBSD" sites are running
the same "FreeBSD" as www.jpr.com.
If you're not sure which "FreeBSD" JP is running, go to Netcraft
and ask about www.jpr.com and then go to www.jpr.com to see
exactly which version he's running.
Tom Podnar
Microlite
Jean-Pierre Radley
Exactly which URL were you looking at, Tom?
--
JP
Joe Dunning
I think his point is that, according to Netcraft, "www.jpr.com" is
running FreeBSB, while the front page of www.jpr.com claims that the
site is running on OpenServer.
Either Netcraft is wrong, or it's about time that you updated your
website in line with the change from OpenServer to FreeBSD about 2 years
ago.
>
>
FyRE
[...]
Now this is interesting. Firstly, Brian "I have the SCO logo stamped
on my ass" White says:
>These quiet readers may get the impression that these wacko's
>statements actually have some validity if no one seems to offer any
>counter-argument, or simply point out the flaws in their arguments.
Which, barring the pointless name-calling states his position. But
then:
(where
>"good service" means they are not susceptible to all the linux
>exploits that keep coming around almost as bad as windows lately, and
An obvious lie. In fact, such an obvious lie that I'm surprised anyone
would say something so stupid in any newsgroup; even one with mostly
SCO flag-wavers (like Brian) as subscribers.
>you don't have to touch their boxes for months or years at a time)
Bear in mind that security issues are not so important for SCO's
archaic products (are they still even supported?) since nobody in
their right mind places a Unixware/openserver on a 'net facing
connection. Even SCO themselves use Linux for their website. Shows how
much confidence they have in their former product line, you know, back
when they used to write software...
>And frankly, I don't need any more people in the world forming
>backwards and counter-productive attitudes based on the bad info some
>people fling around
"Hello Mr Kettle, you're black!"
>Every time some idiot convinces a customer they should be running
>their main application on Windows, it makes my life harder because I then have to
>not only educate them about all the obvious and subtle and direct and
>indirect ways in which relying on unix is more sane than relying on
>windows, I also have to de-educate them about the false promises they
>were already given and convinced of.
Windows -> Linux makes a lot of sense. However, Windows -> SCO is
probably the worst move anyone could make. Not only is the company no
longer producing software (since their whole direction is now set by
the disproportionately large legal team), but they have no future in
any endevour. Why convince someone to adopt a dead platform? You might
as well start pushing "BeOS". It'd be amusing to hear your sales
patter though:
Customer: "Why should we move from Windows to SCO?"
Brain: "Erm, well it supports some terminal based software from 1980
and you can run it on a 386sx! SCO only stopped developing their OS's
10 months ago, so they're pretty recent. I've heard that we might even
get PAM authentication one day, once someone figures out how to
recompile the Linux code that was stolen from SCO to do it!"
Customer: "What about hardware support?"
Brain: "Worry not! All SCO's products support motherboards, graphic
cards, SCSI, RAID for hardware over 5 years old! The new drivers for
the Matrox G200 are right around the corner! We're only a few years
behind eveyone else."
Customer: "How about community support?"
Brain: "Uhhh.... Well.... We do take stuff from the Opensource
community..."
Customer: "But isn't SCO trying to kill off OpenSource?"
Brain: "Well... Yes, but until that happens, we can keep taking stuff
and using it in our products. It's win win"
>Generally, people don't _really_ "get it" until their windows server
>farts and bletcherates their data or costs them several thousand bucks
>in anti-virus-related labor and down-time that would have never
>happened on unix. I can't stand it when that happens. It's a waste of
>time, money, effort, and brains, that could really just about be
>considered criminal.
A fair point, and one I make when promoting Linux. However, since
there's virtually no recent software written to run on SCO's
platforms, it doesn't work when pushing their products. "Sure, there's
no virii written for UnixWare, however, there's no actually apps for
it either..."
>(What else do you call it when you make a promise, take the money, and
>then don't deliver, and charge more money to fix problems that were
>promised not to happen in the first place?)
Here's another question: What do you call it when you send out
threatening letters to people, demanding they pay you $699 for every
CPU running someone elses operating system? This, despite no proof you
deserve anything. I'd call it wire fraud.
>To a lesser extent, the
>over-zealous militant linux contingent suffer from the same false
>foundation. Linux is indeed a better foundation than windows, but if
>you try to claim it's an equivalent replacement for a good solid
>commercial unix, you are either lying, or at best merely ignorant of
>the larger real world where stability matters above all else.
More rubbish. Since Linux runs the vast majority of the web, including
Google, as well as Oracle's internal systems, and is increasingly the
"standard" on IBM servers I beg to differ. Oh yes, it's also used in
many of the biggest clustered servers on Earth. I'm also informed that
it's been PROVEN that Linux is taking massive amounts of market share
from all commercial unices, in every area from movie studios, across
the financial industry and even retail (I know for a fact that one of
the UK's biggest retailers have Linux servers in ALL of their stores).
Which manufacturers install Unixware as standard, BTW? What was that?
None you say? ;-)
>(stability over time and across versions, not just stability as in
>long uptimes, although of course that is important too)
What Brain means here, is that SCO's products have barely changed in
10 years. No arguments here.
>So we offer counter-statements pretty much merely to keep the wackos
>voices from being the only voices and to ensure that at the very least
>the counter-proofs are on record.
For "counter-arguments" read "lies".
Bill Vermillion
Bill Campbell <bi...@celestial.com> wrote:
>On Sat, Jan 17, 2004, Bill Vermillion wrote:
>>In article <KQ_Nb.81841$xy6.140563@attbi_s02>,
>>Joe Dunning <joedunning12...@hotmail.com> wrote:
>>>On 16 Jan 2004 15:01:26 -0800, Brian K. White <br...@aljex.com> wrote:
>>>Clearly you have to exclude uptimes, since, if you look for the
>>>longest uptimes on Netcraft, you will see ONLY various BSDs
>>>and a couple of Linux machines. Not a single instance of a
>>>commercial OS.
>>Not exactly true. The top-20 sites have a preponderance of
>>BSD/OS - and THAT is a commercial version based on the original
>>4.4Lite. Only recently has it been dropped by WindRiver - so as a
>>just a couple of months ago there is no commercial BSD/OS left.
>>
>...
>>They are execptionally stable machines and my servers typically go
>>over a year before I upgrade to a newer OS. Only one time in the
>>past two years was there a patch that required a kernel change and
>>a reboot. Other than that it was patch a module, and restart the
>>module with the OS continuing on. Longest I went without reboot
>>for upgraded was about 750 days.
>The limiting factors for most of the machines we support are
>power outages or equipment moves (all of our systems here have
>been up 91 days which was the last power outage before I bought a
>generator that got us through the last outage).
At the small ISP years ago we had instances of the long power
outages that outlasted the UPS systems, so the scavenger of the
group bought a 50KW Liebert at a local auction. Before we got
that installed properly most of the units were moved to another
location and there was a 100KW unit that we could tack our machines
onto.
> One of our customers has a Linux box at the local Sprint CO
>along with their Annex 8000 RACs where the Linux box is used
>primarily for radius authentication, secondary MX forwarding,
>and DNS. This box had been up for about 450 days until Sprint
>decided to reorganize their rack and cables 203 days ago (Caldera
>eDesktop 2.4).
All my current servers and customers are in the rack space we
have in the local Level (3) facililty. The battery room has more
square footage than my house. When there is power failure the
1.25MegaWatt Caterpillar starts up immediately but there is enough
reserve battery power [4-8 hours - not sure how much] to keep
things going until the Cat gets fired up >>IF<< it ever had a
problem. And there's a 6000 gallon tank next to it.
I finally stopped worrying about power years ago. I had one time
with a flaky port on one of their Cisco 12000 switches where things
were slow. So we had about 45 minutes of poor connectivity. That's
damn good for 4 years [this coming month].
After having machine in a carrier faciltiy if given the choice I'd
never go elsewhere.
Bill Vermillion
What is also interesting is that 'uptime' is not available
for many OSes. They can get uptime from BSD/OS, FreeBSD after
4.3, HP-UX, Linux for version after 2.1 thru 2.5.25 on iNTEL
processors, and Linux on some other processors, Solaris 2.6 and
later, Windows 2000, Windows 2003 Server, and Windows XP.
The inconsistancy of Linux in their treatment fits with the other
comment in this thread that Bill Campbell makes that some Linux
things seem to change without warning and others with no or little
backward compatibility.
Not available on AIX, SCO, old Solaris.
It also mentions that Linux, HP-UX, Solaris, and recent releases
of FreeBSD cycle their uptime back to 0 after 497 so they look
like they had been rebooted then.
Jean-Pierre Radley
I understood that to be his point, but I cannot see the string
"www.jpr.com" on any Netcraft pages I've looked at.
--
JP
Joe Dunning
Put "www.jpr.com" into the "what's that site running" box, or try this
link:
http://uptime.netcraft.com/up/graph/?host=www.jpr.com
>
Joe Dunning
wrote:
>
> The inconsistancy of Linux in their treatment fits with the other
So because Linux changed some undocumented, unspecified and unimportant
behavior (presumably in the TCP/IP stack) that is used by Netcraft, that
is an example of "inconsistancy" is it?
OK, so, in upgrading machines, I have had to change configuration files
in order to make things work. It can be a little frustrating if you are
in a hurry to complete the upgrade. Postfix is an example, going from
1.x to 2.x. there were some changes to the configuration that resulted
in some bounced emails. I don't see how this is a "Linux" problem
though. Also, I was happy to make the changes to get some improvements
in the way the programs work.
Bill Campbell
I don't see how postfix changes relate to Linux given that (a) postfix
isn't Linux specific, and (b) the changes I saw in the main.cf file weren't
fatal but just generated warning messages in the postfix log files.
On the other hand, I have a problem with things like adding readline
support to ``bc'' which then echoed back the input broke a bunch of shell
scripts that had been running since 1985 or earlier without a hitch, or
changing the ``ps'' command to fail if one used a ``-'' option unless one
set the I_WANT_A_BROKEN_PS environment variable,
Bill
--
INTERNET: bi...@Celestial.COM Bill Campbell; Celestial Software LLC
UUCP: camco!bill PO Box 820; 6641 E. Mercer Way
FAX: (206) 232-9186 Mercer Island, WA 98040-0820; (206) 236-1676
URL: http://www.celestial.com/
``When dealing with any spammer, one must always keep in mind that you
are dealing with someone who makes their living through forgery, fraud,
theft, subterfuge and obfuscation. Stated simply, spammers lie.''
David Ritz <dr...@primenet.com>
Bill Vermillion
Bill Campbell <bi...@celestial.com> wrote:
>On Sat, Jan 17, 2004, Joe Dunning wrote:
>>On Sat, 17 Jan 2004 16:55:00 GMT, Bill Vermillion <b...@wjv.comREMOVE>
>>wrote:
>>> The inconsistancy of Linux in their treatment fits with the other
>>So because Linux changed some undocumented, unspecified and
>>unimportant behavior (presumably in the TCP/IP stack) that is
>>used by Netcraft, that is an example of "inconsistancy" is it?
Inconsistant among the various verisions of Linux. That was just
one example and I was commenting on the original post of Campbell's
and he did even mention that one. I like consistancy and backward
compatibility in a platform as when you have users on various
versions you can handle them all the same way.
>>OK, so, in upgrading machines, I have had to change
>>configuration files in order to make things work. It can be
>>a little frustrating if you are in a hurry to complete the
>>upgrade. Postfix is an example, going from 1.x to 2.x. there
>>were some changes to the configuration that resulted in some
>>bounced emails. I don't see how this is a "Linux" problem
>>though. Also, I was happy to make the changes to get some
>>improvements in the way the programs work.
>I don't see how postfix changes relate to Linux given that (a)
>postfix isn't Linux specific, and (b) the changes I saw in the
>main.cf file weren't fatal but just generated warning messages in
>the postfix log files.
>On the other hand, I have a problem with things like adding
>readline support to ``bc'' which then echoed back the input broke
>a bunch of shell scripts that had been running since 1985 or
>earlier without a hitch, or changing the ``ps'' command to fail
>if one used a ``-'' option unless one set the I_WANT_A_BROKEN_PS
>environment variable,
That ps tripped me so often. Changes like this may not seem
noticeable to many Linux users but when you go into many user
machines with various operating systems things which have been
standard in Unix since the earliest days just seemed to change.
You learn to live with it but I'd still like not to have to treat
those machines diffently than the rest. It's just more work.
Bill
Jean-Pierre Radley
Thanks. I see their info on www.jpr.com flitting between SCO UNIX, FreeBSD,
and unknown. False, as the site has always been running SCO UNIX.
I suppose I could report it to them, but that doesn't solve the issue
of how dependable any of rest of their information might be.
--
JP
Tony Lawrence
>Windows -> Linux makes a lot of sense. However, Windows -> SCO is
>probably the worst move anyone could make. Not only is the company no
>longer producing software (since their whole direction is now set by
>the disproportionately large legal team), but they have no future in
>any endevour. Why convince someone to adopt a dead platform? You might
>as well start pushing "BeOS". It'd be amusing to hear your sales
>patter though:
You know, you take other people to task (rightfully) for making
dead wrong statements, but you have done the same thing here.
SCO most definitely IS STILL PRODUCING SOFTWARE.
>there's virtually no recent software written to run on SCO's
>platforms, it doesn't work when pushing their products. "Sure, there's
Another incorrect statement.
>no virii written for UnixWare, however, there's no actually apps for
>it either..."
Ditto.
Pot, Kettle ?
--
to...@aplawrence.com Unix/Linux/Mac OS X resources: http://aplawrence.com
Get paid for writing about tech: http://aplawrence.com/publish.html
Tony Lawrence
>On Sat, 17 Jan 2004 16:55:00 GMT, Bill Vermillion <b...@wjv.comREMOVE>
>wrote:
>>
>> The inconsistancy of Linux in their treatment fits with the other
>So because Linux changed some undocumented, unspecified and unimportant
>behavior (presumably in the TCP/IP stack) that is used by Netcraft, that
>is an example of "inconsistancy" is it?
Oh, bull. As much as I love Linux, Bill's statements are valid. Linux
changes like the wind and it's fricking aggravating.
Tony Lawrence
>On Sat, Jan 17, 2004, Joe Dunning wrote:
>>On Sat, 17 Jan 2004 16:55:00 GMT, Bill Vermillion <b...@wjv.comREMOVE>
>>wrote:
>>>
>>> The inconsistancy of Linux in their treatment fits with the other
>>
>>So because Linux changed some undocumented, unspecified and unimportant
>>behavior (presumably in the TCP/IP stack) that is used by Netcraft, that
>>is an example of "inconsistancy" is it?
>>
>>OK, so, in upgrading machines, I have had to change configuration files
>>in order to make things work. It can be a little frustrating if you are
>>in a hurry to complete the upgrade. Postfix is an example, going from
>>1.x to 2.x. there were some changes to the configuration that resulted
>>in some bounced emails. I don't see how this is a "Linux" problem
>>though. Also, I was happy to make the changes to get some improvements
>>in the way the programs work.
>I don't see how postfix changes relate to Linux given that (a) postfix
>isn't Linux specific, and (b) the changes I saw in the main.cf file weren't
>fatal but just generated warning messages in the postfix log files.
>On the other hand, I have a problem with things like adding readline
>support to ``bc'' which then echoed back the input broke a bunch of shell
>scripts that had been running since 1985 or earlier without a hitch, or
>changing the ``ps'' command to fail if one used a ``-'' option unless one
>set the I_WANT_A_BROKEN_PS environment variable,
Which is a good example of the incredible hubris of the people who
did ps, by the way.
Personally, I think bsd ps semantics are "broken", not the other
way around. But if I were putting in such a flag, I'd call it
"I_WANT_BSD_FLAGS".
Joe Dunning
It is interesting. nmap reports your site running some form of
Openserver -- it wan't too clear, but my firewall was interfering with
the scan.
The only question could be: is there some kind of caching or proxying
going on? That can fool Netcraft, so that you see combinations like IIS
on Linux.
>
Jean-Pierre Radley
| >|
| >| Put "www.jpr.com" into the "what's that site running" box, or try this
| >| link:
| >| http://uptime.netcraft.com/up/graph/?host=www.jpr.com
| >
| >Thanks. I see their info on www.jpr.com flitting between SCO UNIX, FreeBSD,
| >and unknown. False, as the site has always been running SCO UNIX.
| >
| >I suppose I could report it to them, but that doesn't solve the issue
| >of how dependable any of rest of their information might be.
|
| It is interesting. nmap reports your site running some form of
| Openserver -- it wan't too clear, but my firewall was interfering with
| the scan.
|
| The only question could be: is there some kind of caching or proxying
| going on? That can fool Netcraft, so that you see combinations like IIS
| on Linux.
Nope, none of that.
--
JP
Charlie Gibbs
br...@aljex.com (Brian K. White) writes:
>Generally, people don't _really_ "get it" until their windows server
>farts and bletcherates their data or costs them several thousand bucks
>in anti-virus-related labor and down-time that would have never
>happened on unix.
Unfortunately, many of them don't get it even then - they just shrug
their shoulders, re-boot, re-format, re-install (whatever it takes)
and accept it all as the cost of doing business. And Microsoft is
quite happy to have it that way because they can promise that the
next release really fixes the problem, yessiree. And it's all the
better if users aren't aware that it can ever be any better.
>I can't stand it when that happens. It's a waste of time, money,
>effort, and brains, that could really just about be considered
>criminal.
"Crimes against humanity" is what I prefer to call it. My theory
is that Bill Gates is from another planet. Back in the '60s, his
people observed our rapid technological progress, and became alarmed
that we might soon expand into space and become a threat to them.
So they sent Bill to Earth, and thanks to his efforts, much of the
energy that could have gone into technological progress has instead
been dissipated on re-installing Windows.
--
/~\ cgi...@kltpzyxm.invalid (Charlie Gibbs)
\ / I'm really at ac.dekanfrus if you read it the right way.
X Top-posted messages will probably be ignored. See RFC1855.
/ \ HTML will DEFINITELY be ignored. Join the ASCII ribbon campaign!
Joe Dunning
>In article <60bd4c6b.04011...@posting.google.com>
>br...@aljex.com (Brian K. White) writes:
>
>>Generally, people don't _really_ "get it" until their windows server
>>farts and bletcherates their data or costs them several thousand bucks
>>in anti-virus-related labor and down-time that would have never
>>happened on unix.
>
>Unfortunately, many of them don't get it even then - they just shrug
>their shoulders, re-boot, re-format, re-install (whatever it takes)
>and accept it all as the cost of doing business. And Microsoft is
>quite happy to have it that way because they can promise that the
>next release really fixes the problem, yessiree.
I believe that the 3 great lies are:
1. I'll respect you in the morning
2. The check is in the mail
3. It's fixed in the next release
There is a 4th great lie -- one much used by software companies also:
4. "No-one else has reported that as a problem/bug"
>
>>I can't stand it when that happens. It's a waste of time, money,
>>effort, and brains, that could really just about be considered
>>criminal.
>
>"Crimes against humanity" is what I prefer to call it. My theory
>is that Bill Gates is from another planet. Back in the '60s, his
>people observed our rapid technological progress, and became alarmed
>that we might soon expand into space and become a threat to them.
>So they sent Bill to Earth, and thanks to his efforts, much of the
>energy that could have gone into technological progress has instead
>been dissipated on re-installing Windows.
I don't like MS's behavior in operating vastly different prices
for the same MS software in different countries. What is happening is
that MS is adding to the cost of doing business in western and
developed countries, while selling MS software at 90% (or
more) discounted prices elsewhere (I'm not discussing illegal copies --
just MS's pricing structure).
But what is interesting is that there is a generation of people that
think regular re-installations of OS-es are necessary, random crashes
are to be expected and that machines should be re-booted routinely as
some kind of preventative measure. Really, that's MS's greates
achievement: setting people's expectations for software to be so low!
Tony Lawrence
>I don't like MS's behavior in operating vastly different prices
>for the same MS software in different countries. What is happening is
>that MS is adding to the cost of doing business in western and
>developed countries, while selling MS software at 90% (or
>more) discounted prices elsewhere (I'm not discussing illegal copies --
>just MS's pricing structure).
They are doing the same thing here. A company I have a Linux mailserver
at was quoted $1,600.00 for Exchange for 500 users! The reseller classified
them as a "charitable" org (they are not) in order to get this price. At
that price, I might lose the business..
Bill Campbell
>Joe Dunning <j...@blahblah.invalid> wrote:
>
>>I don't like MS's behavior in operating vastly different prices
>>for the same MS software in different countries. What is happening is
>>that MS is adding to the cost of doing business in western and
>>developed countries, while selling MS software at 90% (or
>>more) discounted prices elsewhere (I'm not discussing illegal copies --
>>just MS's pricing structure).
>
>They are doing the same thing here. A company I have a Linux mailserver
>at was quoted $1,600.00 for Exchange for 500 users! The reseller classified
>them as a "charitable" org (they are not) in order to get this price. At
>that price, I might lose the business..
One must remember the hidden costs of running Exchange servers, rebuilding
them from scratch several times a year when they corrupt their databases,
the constant flood of security issues, etc.
Have you looked at SuSE's exchange replacement systems?
Bill
--
INTERNET: bi...@Celestial.COM Bill Campbell; Celestial Software LLC
UUCP: camco!bill PO Box 820; 6641 E. Mercer Way
FAX: (206) 232-9186 Mercer Island, WA 98040-0820; (206) 236-1676
URL: http://www.celestial.com/
``Capitalism works primarily because most of the ways that a company can be
scum end up being extremely bad for business when there's working
competition.'' -rra
Scott Burns
>To a lesser extent, the
>over-zealous militant linux contingent suffer from the same false
>foundation. Linux is indeed a better foundation than windows, but if
>you try to claim it's an equivalent replacement for a good solid
>commercial unix, you are either lying, or at best merely ignorant of
>the larger real world where stability matters above all else.
>
>
The servers I work with require to go at least six months (sugar cane
crushing season) and can only be rebooted on specified days when the
mill is down for maintenance. The other six months of the year, well,
who cares? So long as the OS can make it that far, only our preference
as the software supplier matters. In out case, it's not a sliding scale
of which OS is better, it's a simple tick or cross to say wether it is
good enough. In our case, Linux and SCO are equally reliable, at least
in theory, and our justifications for one over the other come from
elsewhere.
Now, am I lying, or just plain ignorant? I suspect I'm not involved in
your larger real world. I also suspect that more people are in the
"tick or cross" world than the "sliding scale" world.
--
Scott Burns
Mirrabooka Systems
Tel +61 7 3857 7899
Fax +61 7 3857 1368
Scott Burns
I missed part of your quote. It should end :
|(stability over time and across versions, not just stability as in
|long uptimes, although of course that is important too)
My experience is that machines are deployed, and then 2-5 years later a
replacement is installed from scratch. In the meantime, no updates or
patches are applied to the original. This does not cause a problem as
generally the small vendors who produce 95% of the custom apps are in
the same boat.
My experience is that stability over time is acheived through finacial
constraints and people not upgrading, and has little to do with how much
the underlying system has changed since it was installed. When the
vendor finally does do an upgrade the custom app is likely to have
changed far more than the system on which it resides.
We keep a RH5.2 box around simply because anything compiled on it will
run pretty much anywhere up to and including RH7.2, and probably up to
RH9 (our main machine is RH7.2 so it's easier to just compile on this
for those clients 7.2 and above). RH5.2 to RH7.2 was about 3 years.
7.2 to 9 was about 18 months more. How exactly do we not have
stability over time and versions? Because a couple of shell scripts
need tweaking? (we have few like this, YMMV) In our case it is stable
over the likely life of the machine, so it gets the yes box ticked.
Essentially, I believe Linux is "stable" for most people and
applications. It is a small subset which needs the type of stability
given by the higher priced UNIX varieties.
Bill Vermillion
<a...@shell01.TheWorld.com> wrote:
>Joe Dunning <j...@blahblah.invalid> wrote:
>>I don't like MS's behavior in operating vastly different prices
>>for the same MS software in different countries. What is
>>happening is that MS is adding to the cost of doing business in
>>western and developed countries, while selling MS software at
>>90% (or more) discounted prices elsewhere (I'm not discussing
>>illegal copies -- just MS's pricing structure).
>They are doing the same thing here. A company I have a Linux
>mailserver at was quoted $1,600.00 for Exchange for 500 users!
>The reseller classified them as a "charitable" org (they are
>not) in order to get this price. At that price, I might lose the
>business..
I've talked to one local MS shop who can't sell the MS software
because the edcational verisons are being sold locally in the
electronic discounters and such places as Costco. That means the
prices are far below what the dealer pays.
It's never good when a vendor competew with it's dealers but it
sounds like that's what's happening to you. There should be no
legal way to reclassify a business to get a price break to compete
against other solutions. It's sounds so much like some of the
things MS was being sued for in the past.
Tony Lawrence
> Which is a good example of the incredible hubris of the people who
> did ps, by the way.
>
> Personally, I think bsd ps semantics are "broken", not the other
> way around. But if I were putting in such a flag, I'd call it
> "I_WANT_BSD_FLAGS".
And I was wrong. For sure about "ps" and probably about other stuff.
See "I_WANT_A_BROKEN_PS" at http://aplawrence.com/Blog/B1011.html
and "How "ps" works and why" at http://aplawrence.com/Blog/B1012.html
--
Tony Lawrence
Bill Vermillion
The man page for BSD's ps specifically states that for historical
reaons it supports a different set of options than the POSIX
standards and what is supported on non-BSD operating systems.
The BSD seems to support more options than almost anyone would
desire, and the -O or -o options have 81 different keywords .
It's not broken in BSD-land - it's just different. :-)
Joe Dunning
to sue SCO, just after SCO's case against Daimler-Chrysler (one the 2
"Linux end-user lawsuits") was gutted.
So, I guess, for SCO, that's one case over and a new one beginning
(unless SCO's management want to pursue claims regarding DC's failure to
reply within 30 days).
Still not a shred of credible evidence from SCO regarding SCO's cases.
Now, as Tony pointed out SCO is claiming ownership of a specification
that SCO's predecessors helped make available!
Tony, many months ago, you suggested that SCO might have a case -- that
SCO might have evidence up their sleeve, waiting for the appropriate
moment to come out.
What's your opinion now? Do you still think there is any likelyhood that
SCO has any credible evidence?
Tony Lawrence
>
> Tony, many months ago, you suggested that SCO might have a case --
that
> SCO might have evidence up their sleeve, waiting for the appropriate
> moment to come out.
>
> What's your opinion now? Do you still think there is any likelyhood
that
> SCO has any credible evidence?
No matter how many times I explain it, using the simplest possible
language,
you still don't get it.
I said they "might". I also said that I had no particular knowledge of
any such evidence, probably wouldn't understand it if I tripped over
it, etc. Over and over I explained that this was simple supposition.
I STILL don't know if they have any evidence. I do shake my head in
wonder of this latest Sys V init and ELF crapola ( see
http://aplawrence.com/Blog/B992.html ), but I have neither knowledge or
understanding of whatever legal claims are involved there.
Again: nothing has changed. They might have evidence, they might not.
They might have contracts, copyrights, ironclad proof and so on or they
might have diddly. Whatever they have or don't have, whether they are
morally right or morally wrong, they may win or they may lose. It's
no different now than it was this time last year.
Nor has anything has swayed my opinion as to the damage this will do to
everyone, including their own selves, should they actually win this. I
think the whole thing was a very bad idea and that the only real winner
in any outcome is Microsoft: one way or another, another competitor
gets hurt.
But you'll go on pretending that I've said something different, right,
Joe? You'll pretend that I said "SCO has a case" or something equally
ludicrous. That's what you always do, don't disappoint me now. And
the little parrot caled FyRe will join right in, right?
Gaad - just responding to you leaves a bad taste in my mouth.
--
Tony Lawrence
http://aplawrence.com/Lawsuit/
Joe Dunning
> Joe Dunning wrote:
>
>>
>> Tony, many months ago, you suggested that SCO might have a case --
> that
>> SCO might have evidence up their sleeve, waiting for the appropriate
>> moment to come out.
>>
>> What's your opinion now? Do you still think there is any likelyhood
> that
>> SCO has any credible evidence?
>
> No matter how many times I explain it, using the simplest possible
> language,
> you still don't get it.
Oh, I get it alright. I just find your continuing "three wise monkeys"
act hillarious!
>
> Again: nothing has changed. They might have evidence, they might not.
> They might have contracts, copyrights, ironclad proof and so on or they
> might have diddly. Whatever they have or don't have, whether they are
> morally right or morally wrong, they may win or they may lose. It's
> no different now than it was this time last year.
Now that is manifestly untrue. Since last year, SCO has made claims at
SCOForum, which were quickly debunked; SCO has made claims regarding
ABI, which were debunked; SCO has made claims regarding ELF, which were
quickly debunked. In fact, SCO has done everything in its power
(including making conflicting statements to various courts) to avoid
showing any real evidence. Furthermore, Novell has challenged SCO's
claims to own SVR4 copyrights. So, your opinion or rather your lack of
opinion may not have changed, but let's not pretend things are now the
same as they were this time last year.
> But you'll go on pretending that I've said something different, right,
> Joe? You'll pretend that I said "SCO has a case" or something equally
> ludicrous. That's what you always do, don't disappoint me now. And
> the little parrot caled FyRe will join right in, right?
No, I'm not going to claim that you said SCO has a case. I do understand
the difference between "have" and "might have", but it's obvious you are
following the case on some level, sonce you picked up on the ELF claims
pretty quickly.
I'm just amused that you can claim to have no opinion on the strength or
otherwise of SCO's case given all the information that has come out in
the last year.
Tony Lawrence
All true. And all hopeful signs. But still nothing has changed. They
could still win this, or cause problems even by losing. But, as I've
said all along, there are much more important things afoot. This is a
part of the threats to Linux, Open Source et al., but hardly the most
important part.
>
> > But you'll go on pretending that I've said something different,
right,
> > Joe? You'll pretend that I said "SCO has a case" or something
equally
> > ludicrous. That's what you always do, don't disappoint me now.
And
> > the little parrot caled FyRe will join right in, right?
>
> No, I'm not going to claim that you said SCO has a case. I do
understand
> the difference between "have" and "might have", but it's obvious you
are
> following the case on some level, sonce you picked up on the ELF
claims
> pretty quickly.
Oh, *finally* you understand? I'll have to frame this one!
>
> I'm just amused that you can claim to have no opinion on the strength
or
> otherwise of SCO's case given all the information that has come out
in
> the last year.
Once again, little man, I am not a lawyer, and I don't have intimate
knowledge of Unix history or the various contracts and agreements
involved here. I'm sure there are a few people truly qualified to have
an opinion in these matters, but I doubt we've heard from any of them
here or at GrokLaw.
I can look at the facts and wonder how in hell ELF could possibly be
claimed by SCO, but that's no different than my wondering how in hell
the U.S. Supreme court has ruled that a U.S. citizen can be held
indefinitely without trial or even being charged. Both things may
astound me, but I'm not qualified to argue about either.
And then, of course, there's the people factor. I might guess how David
Souter will vote on a particlar issue based on what I know of his
politics, but he can surprise me, With SCO's suits, I know nothing
about the various judges, their general intelligence, their tech
background, their political leanings, etc. so how on earth could I even
begin to estimate what they might think of evidence I haven't seen?
It would be ludicrous for me to have an opinion on this. The hilarious
part is that we both know that if I *did* have an opinion you didn't
like, you and the other jackasses would be jumping up and down saying
all the things I have said in the previous paragraphs.
I have hopes about the outcome, not opinions. Further, my hopes for
this are just part of a larger yearning for limiting corporate power
and allowing room for innovation. My hopes have been slightly buoyed
by some developments here, but dampened by other occurences that you'd
probably see as unrelated.
"Being in politics is like coaching football. You have to be smart
enough to understand the game and dumb enough to think it's important."
- Eugene J. McCarthy
Reminds me of you, for some reason.
tha...@dexter.glaci.com
>
> I can look at the facts and wonder how in hell ELF could possibly be
> claimed by SCO, but that's no different than my wondering how in hell
> the U.S. Supreme court has ruled that a U.S. citizen can be held
> indefinitely without trial or even being charged. Both things may
> astound me, but I'm not qualified to argue about either.
OK, I usually just lurk when these arguments are going on, but I
just have to jump in on this point. You say that your are not
qualified to argue about these issues of the law and constitution.
If you are a citizen with a brain and a pulse, you are not only
qualified, you are *obligated* to have a point of view and to
act on it. A constitutional democracy can not function without
the involvement of an informed electorate... otherwise it
devolves into a sham democracy being controlled by a greedy
corrupt elite (please hold back your shouts that it has already
happened).
This idea that the constitution and the law is some sort of
lofty domain that only lawyers and politicians are qualified to
argue about is BULLSHIT! The constitution was written to be
a document that all Americans could read and understand. It
is the role of the court system and our political leaders to
make sure that our rights and freedoms under that document are
respected, and it is our right and DUTY as citizens to hold
their feet to the fire when we see them trampling on the
constitution. Lawyers play an important role in clarifying
issues and translating complex legal language into language
us common folks can understand; they represent their clients
and help them navigate the labyrinth that is our legal system,
but they are not the owners of the law... we the citizens are!
If you are astounded by our government's behavior and think the
Constitution is being ignored or undermined... SPEAK OUT! Don't
tell yourself that you are not qualified. Your qualifications
are inherent in your citizenship. If you feel you don't have
all the facts, investigate and learn them. If you think that
all the facts are not available... then speak up about that.
Demand that the facts be made available.
If we as citizens ever decide that we are not qualified to
understand the laws under which we are required to live, that
we are not capable of participating in our own governance...
then we have traded our citizenship and our very freedom for
the bonds of slavery.
I'll wipe the froth from my chin and jump off my soap box now,
but gosh darn that sort of thinking does get me riled up.
P.S. Don't forget to vote.
--
Thad Phetteplace - GLACI, Inc.
http://www.GridSlammer.org - An open source video game toolkit
Joe Dunning
> Joe Dunning wrote:
>> On 2004-07-25, Tony Lawrence <pcu...@gmail.com> wrote:
>> > Joe Dunning wrote:
>
>
>> >
> Once again, little man, I am not a lawyer, and I don't have intimate
> knowledge of Unix history or the various contracts and agreements
> involved here. I'm sure there are a few people truly qualified to have
> an opinion in these matters, but I doubt we've heard from any of them
> here or at GrokLaw.
Once again you fail to differentiate between *having an opinion* and
*whether that opinion matters*. I do have opinions on many subjects. Do
I think anyone else gives a damn about them? Of course not. Do I think
my opinion will change anything? No. Let's face it: the only people
whose opinions matter are the various judges and (if it ever gets that
far) the juries, and possibly appeals court judges. No-one else's
opinions really matter. Not even the lawyers. Qualification is not
relevant to whether your opinion matters.
Incidentally, that does bring an interesting thought: according to your
logic, we should abandon the jury system, since juries are not qualified
to hold opinions on most subjects.
What's more, in your twisted imaginings of my thinking, you probably
think that I always thought SCO has no case: not true. My initial
reaction was the same as yours: they might have. It is only because of
the information that has come out over the past year, the wild claims
and lack of any credible evidence from SCO that my opinion has been
formed -- oh, and reading the various licensing agreements. In other
words: my opinion is based on the facts available to me.
>
> I can look at the facts and wonder how in hell ELF could possibly be
> claimed by SCO, but that's no different than my wondering how in hell
> the U.S. Supreme court has ruled that a U.S. citizen can be held
> indefinitely without trial or even being charged. Both things may
> astound me, but I'm not qualified to argue about either.
>
> And then, of course, there's the people factor. I might guess how David
> Souter will vote on a particlar issue based on what I know of his
> politics, but he can surprise me, With SCO's suits, I know nothing
> about the various judges, their general intelligence, their tech
> background, their political leanings, etc. so how on earth could I even
> begin to estimate what they might think of evidence I haven't seen?
>
> It would be ludicrous for me to have an opinion on this. The hilarious
> part is that we both know that if I *did* have an opinion you didn't
> like, you and the other jackasses would be jumping up and down saying
> all the things I have said in the previous paragraphs.
No: the situation is entirely different: I don't have an opinion on the
subject because I have not studied it. On the other hand the SCO
sitation: I have studied, I note that you have also followed it over the
past year. Even so, were you to have an opinion, I recognize your right
to HOLD an opinion -- again, the context for such an opinion is that if
either of us held one, it probably would not affect anything.
>
> I have hopes about the outcome, not opinions. Further, my hopes for
> this are just part of a larger yearning for limiting corporate power
> and allowing room for innovation. My hopes have been slightly buoyed
> by some developments here, but dampened by other occurences that you'd
> probably see as unrelated.
>
> "Being in politics is like coaching football. You have to be smart
> enough to understand the game and dumb enough to think it's important."
> - Eugene J. McCarthy
>
> Reminds me of you, for some reason.
Once again Tony, you resort to name-calling!