Sun Linux desktops
John Miller
tuca...@whodis.org
Tut tut, it looks like rain.
Chris Newport
>
> http://www.linuxjournal.com/article.php?sid=6327&mode=thread&order=0
DO NOT FEED THE TROLL.
Rich Teer
> http://www.linuxjournal.com/article.php?sid=6327&mode=thread&order=0
Bah. The only Linux on The Desktop (LOTD) I'll have is
the old Red Hat CD I use as a coaster...
--
Rich Teer
President,
Rite Online Inc.
Voice: +1 (250) 979-1638
URL: http://www.rite-online.net
tuca...@whodis.org
: the old Red Hat CD I use as a coaster...
Lol!
yalu
>> http://www.linuxjournal.com/article.php?sid=6327&mode=thread&order=0
>
> DO NOT FEED THE TROLL.
It's funny to see how this is being interpreted as troll. What's wrong?
Sun will never abandon Solaris in favour of GNU/Linux, reading this
newsgroup I know the users simply won't allow it :-)
It's not that Sun had a deal with Microsoft to port Xp to the sparc. THAT
would've been quiet a story.
P.S. yes of course it's troll, because it was destined to be interpreted
as such.
--
mail: frankpuntvandammebijstudent-kuleuven-ac-belgie
homepage: www.student.kuleuven.ac.be/~m9917684
jabber: ya...@jabber.com
I R A Darth Aggie
Chris Newport <c...@NOSPAM.netunix.com>, in
<3D87CD7B...@NOSPAM.netunix.com> wrote:
+> John Miller wrote:
+> >
+> > http://www.linuxjournal.com/article.php?sid=6327&mode=thread&order=0
+>
+> DO NOT FEED THE TROLL.
Ok, but this isn't a troll.
Would someone please explain to me Sun's linux strategy? if you'd read
the posted link, you'd see this
<begin quote>
The announcement is still under wraps, although Don Clark of the Wall
Street Journal got a look under the shroud and found what he saw
compelling enough to warrant yet another Threat-to-Microsoft story. It
ran Thursday on the front page of the Marketplace section, as one of
the lead stories.
For clues about where this thing is going to fit, here's the key
paragraph from the Journal piece:
In the biggest development, Microsoft archival Sun
Microsystems Inc. next week plans to announce its first
full-fledged commitment to Linux on desktop PCs. The
computer maker, whose server business has been hurt by
low-end systems running Microsoft Windows, is determined to
counterattack by cutting into Microsoft's cash-cow franchise
in desktop-PC software. "We have a chance to be a force for
change in the industry," says Jonathan Schwartz, Sun's
executive vice president of software.
<end quote>
If you want to claim that the WSJ is FUD Central, by all means do try.
James
--
Consulting Minister for Consultants, DNRC
I can please only one person per day. Today is not your day. Tomorrow
isn't looking good, either.
I am BOFH. Resistance is futile. Your network will be assimilated.
Jonathan Buzzard
yalu <frank...@mail.com> writes:
> On Wed, 18 Sep 2002 02:48:59 +0200, Chris Newport wrote:
>
>>> http://www.linuxjournal.com/article.php?sid=6327&mode=thread&order=0
>>
>> DO NOT FEED THE TROLL.
>
> It's funny to see how this is being interpreted as troll. What's wrong?
> Sun will never abandon Solaris in favour of GNU/Linux, reading this
> newsgroup I know the users simply won't allow it :-)
Never is a very long time indeed. One should never say never as you
invariably get proved wrong in the long run. I once upon a time heard
similar views expressed about Sun never dropping Sun OS 4.1.x and we
all know what happened there...
It would be a *lot* safer to say that Sun won't be abandoning Solaris
in favour of Linux any time soon, but 10 years from now who knows.
JAB.
--
Jonathan A. Buzzard Email: jona...@buzzard.org.uk
Northumberland, United Kingdom. Tel: +44(0)1661-832195
Chuck Swiger
[ ... ]
> Never is a very long time indeed. One should never say never as you
> invariably get proved wrong in the long run.
One should almost never say "never", rather.
Otherwise, your statement itself would be a self-refuting tautology. :-)
-Chuck
Chuck Swiger | ch...@codefab.com | All your packets are belong to us.
-------------+-------------------+-----------------------------------
"The human race's favorite method for being in control of the facts
is to ignore them." -Celia Green
yalu
> Never is a very long time indeed. One should never say never as you
> invariably get proved wrong in the long run. I once upon a time heard
> similar views expressed about Sun never dropping Sun OS 4.1.x and we
> all know what happened there...
Maybe I look extremely dumb now, but isn't Solaris a descendent of SunOS?
> It would be a *lot* safer to say that Sun won't be abandoning Solaris in
> favour of Linux any time soon, but 10 years from now who knows.
That's nearly the eternity in the computer world...
Who knows what gnu/linux can do at that point?
Anyway, I don't think the next two major kernel releases won't replace
Solaris on high-end machines. Does that sound more careful? :-)
I R A Darth Aggie
yalu <frank...@mail.com>, in
<pan.2002.09.18.1...@mail.com> wrote:
+> On Wed, 18 Sep 2002 18:21:12 +0200, Jonathan Buzzard wrote:
+>
+> > Never is a very long time indeed. One should never say never as you
+> > invariably get proved wrong in the long run. I once upon a time heard
+> > similar views expressed about Sun never dropping Sun OS 4.1.x and we
+> > all know what happened there...
+>
+> Maybe I look extremely dumb now, but isn't Solaris a descendent of SunOS?
Not according to my understanding. SunOS was descended from BSD, and
Solaris is a SysV variant. Which is not to deny that there are BSD
and/or SunOS bits in Solaris...
However, I wasn't around Sun boxen during the transition, so I can't
tell you how many bits of SunOS have been retained. I'm guessing that
would have been around 1992 or 1993.
+> Anyway, I don't think the next two major kernel releases won't replace
+> Solaris on high-end machines. Does that sound more careful? :-)
Yes.
YTC#1
<jona...@buzzard.org.uk> wrote:
> In article <pan.2002.09.18.10...@mail.com>,
> yalu <frank...@mail.com> writes:
>> On Wed, 18 Sep 2002 02:48:59 +0200, Chris Newport wrote:
>>
>>>> http://www.linuxjournal.com/article.php?sid=6327&mode=thread&order=0
>>>
>>> DO NOT FEED THE TROLL.
>>
>> It's funny to see how this is being interpreted as troll. What's wrong? Sun
>> will never abandon Solaris in favour of GNU/Linux, reading this newsgroup I
>> know the users simply won't allow it :-)
>
> Never is a very long time indeed. One should never say never as you invariably
> get proved wrong in the long run. I once upon a time heard similar views
> expressed about Sun never dropping Sun OS 4.1.x and we all know what happened
> there...
>
> It would be a *lot* safer to say that Sun won't be abandoning Solaris in
> favour of Linux any time soon, but 10 years from now who knows.
I think the Solaris timeline is 15 years.
--
Bruce Porter
XJR1300SP, XJ900F, GSX750W, GS550, GSX250, CB175
POTM#1, WUSS#1 , YTC#1(bar), OSOS#2 , DS#3 , IbW#18 ,Apostle#8
"The internet is a huge and diverse community
and not every one is friendly"
http://www.ytc1.co.uk
I R A Darth Aggie
Bryan Althaus <br...@panix.com.invalid>, in
<ambfbr$50a$1...@reader1.panix.com> wrote:
+> Sun dropped SunOs 4.x because the code base had become a little worn and
+> tacking on things like SMP, soft real-time and kernel threads would just
+> have been to much. So Solaris was born.
Also, wasn't AT&T making noises about their unlicensed code being used
in the commercial BSD derivatives? the typical ones: money, lawsuits,
lawyers?
Jonathan Buzzard
"YTC#1" <y...@ytc1.co.uk> writes:
> In article <o59ama...@192.168.42.254>, "Jonathan Buzzard"
> <jona...@buzzard.org.uk> wrote:
>
>> In article <pan.2002.09.18.10...@mail.com>,
>> yalu <frank...@mail.com> writes:
>>> On Wed, 18 Sep 2002 02:48:59 +0200, Chris Newport wrote:
>>>
>>>>> http://www.linuxjournal.com/article.php?sid=6327&mode=thread&order=0
>>>>
>>>> DO NOT FEED THE TROLL.
>>>
>>> It's funny to see how this is being interpreted as troll. What's wrong? Sun
>>> will never abandon Solaris in favour of GNU/Linux, reading this newsgroup I
>>> know the users simply won't allow it :-)
>>
>> Never is a very long time indeed. One should never say never as you invariably
>> get proved wrong in the long run. I once upon a time heard similar views
>> expressed about Sun never dropping Sun OS 4.1.x and we all know what happened
>> there...
>>
>> It would be a *lot* safer to say that Sun won't be abandoning Solaris in
>> favour of Linux any time soon, but 10 years from now who knows.
>
> I think the Solaris timeline is 15 years.
>
I think you might find it is a lot closer than that.
YTC#1
<jona...@buzzard.org.uk> wrote:
> In article <20020918.214452...@ytc1.co.uk>,
> "YTC#1" <y...@ytc1.co.uk> writes:
><snip>
>>>
>>> It would be a *lot* safer to say that Sun won't be abandoning Solaris in
>>> favour of Linux any time soon, but 10 years from now who knows.
>>
>> I think the Solaris timeline is 15 years.
>>
>>
> I think you might find it is a lot closer than that.
I mean the dev cycle, ie how far ahead it is being looked at.
bit-b...@maney.org
: On Wed, 18 Sep 2002 18:21:12 +0200, Jonathan Buzzard wrote:
:> Never is a very long time indeed. One should never say never as you
:> invariably get proved wrong in the long run. I once upon a time heard
:> similar views expressed about Sun never dropping Sun OS 4.1.x and we
:> all know what happened there...
: Maybe I look extremely dumb now, but isn't Solaris a descendent of SunOS?
No. SunOS is the operating system. Solaris is the operating environment.
Basically Solaris is SunOS (the operating system) + a slew of applications
that give a certain "look-n-feel" (the operating environment).
:> It would be a *lot* safer to say that Sun won't be abandoning Solaris in
:> favour of Linux any time soon, but 10 years from now who knows.
: That's nearly the eternity in the computer world...
: Who knows what gnu/linux can do at that point?
Far less than linux will do for sure.
: Anyway, I don't think the next two major kernel releases won't replace
: Solaris on high-end machines. Does that sound more careful? :-)
Linux won't be creplacing Solaris in my environment anytime soon,
on any of my machines, regardless of what the market droids and hype
from the Street are saying.
fpsm
--
| Fredrich P. Maney my_last_name AT my_last_name DOT org |
| Do NOT send me HTML formatted E-mail or copies of netnews posts! |
| Address in header is a spamtrap. Use one in signature for replies. |
| Please review http://www.maney.org/uce/ before emailing. |
Jonathan Buzzard
bit-b...@maney.org writes:
> In comp.sys.sun.misc yalu <frank...@mail.com> wrote:
>: On Wed, 18 Sep 2002 18:21:12 +0200, Jonathan Buzzard wrote:
>
>:> Never is a very long time indeed. One should never say never as you
>:> invariably get proved wrong in the long run. I once upon a time heard
>:> similar views expressed about Sun never dropping Sun OS 4.1.x and we
>:> all know what happened there...
>
>: Maybe I look extremely dumb now, but isn't Solaris a descendent of SunOS?
>
> No. SunOS is the operating system. Solaris is the operating environment.
> Basically Solaris is SunOS (the operating system) + a slew of applications
> that give a certain "look-n-feel" (the operating environment).
Not really the internals of the SunOS 4.1.x (aka Solaris 1.x) and
Solaris 2.x onwards are entirely different. One (SunOS 4.1.x) is
a BSD Unix and the other is a SysV Unix.
>
> Linux won't be creplacing Solaris in my environment anytime soon,
> on any of my machines, regardless of what the market droids and hype
> from the Street are saying.
>
Like I said, that is what people said about Solaris 2.x, They would
not be replacing there SunOS 4.1.x machines with this new fangled
Solaris 2.x
If in a couple of years time Sun decide to drop support and development
for Solaris you will have little to no choice but to follow any switch.
Steven Shelikoff
We still run SunOS 4.1.x on a lot of machines.
Steve
bit-b...@maney.org
: bit-b...@maney.org writes:
[...]
:> No. SunOS is the operating system. Solaris is the operating environment.
:> Basically Solaris is SunOS (the operating system) + a slew of applications
:> that give a certain "look-n-feel" (the operating environment).
: Not really the internals of the SunOS 4.1.x (aka Solaris 1.x) and
: Solaris 2.x onwards are entirely different. One (SunOS 4.1.x) is
: a BSD Unix and the other is a SysV Unix.
It's still SunOS with the OE on top of it that is called Solaris.
Whether or not the Sun OS under it is BSD or SysV based is completely
irrelevant in this context.
:> Linux won't be creplacing Solaris in my environment anytime soon,
:> on any of my machines, regardless of what the market droids and hype
:> from the Street are saying.
: Like I said, that is what people said about Solaris 2.x, They would
: not be replacing there SunOS 4.1.x machines with this new fangled
: Solaris 2.x
: If in a couple of years time Sun decide to drop support and development
: for Solaris you will have little to no choice but to follow any switch.
No. As I said, Linux won't be replacing Solaaris on my machines. If
Sun does force me to drop Solaris, I'll be going to BSD, not Linux.
And if that means dropping Sun hardware as well, well that will be
their own damn fault, won't it?
bit-b...@maney.org
: In comp.unix.solaris Jonathan Buzzard <jona...@buzzard.org.uk> wrote:
[...]
: : Never is a very long time indeed. One should never say never as you
: : invariably get proved wrong in the long run. I once upon a time heard
: : similar views expressed about Sun never dropping Sun OS 4.1.x and we
: : all know what happened there...
: Sun dropped SunOs 4.x because the code base had become a little worn and
: tacking on things like SMP, soft real-time and kernel threads would just
: have been to much. So Solaris was born.
Wrong. SunOS 5.x was born. SunOS is the operating system. Solaris is
the Operating System + the Operating Environment.
[...]
Jonathan Buzzard
bit-b...@maney.org writes:
> In comp.sys.sun.misc Jonathan Buzzard <jona...@buzzard.org.uk> wrote:
>: In article <amd0ta$de7$1...@newsread.stdio.com>,
>: bit-b...@maney.org writes:
>
> [...]
>
>:> No. SunOS is the operating system. Solaris is the operating environment.
>:> Basically Solaris is SunOS (the operating system) + a slew of applications
>:> that give a certain "look-n-feel" (the operating environment).
>
>: Not really the internals of the SunOS 4.1.x (aka Solaris 1.x) and
>: Solaris 2.x onwards are entirely different. One (SunOS 4.1.x) is
>: a BSD Unix and the other is a SysV Unix.
>
> It's still SunOS with the OE on top of it that is called Solaris.
> Whether or not the Sun OS under it is BSD or SysV based is completely
> irrelevant in this context.
Not really, the point is that a kernel change was forced on you.
>:> Linux won't be creplacing Solaris in my environment anytime soon,
>:> on any of my machines, regardless of what the market droids and hype
>:> from the Street are saying.
>
>: Like I said, that is what people said about Solaris 2.x, They would
>: not be replacing there SunOS 4.1.x machines with this new fangled
>: Solaris 2.x
>
>: If in a couple of years time Sun decide to drop support and development
>: for Solaris you will have little to no choice but to follow any switch.
>
> No. As I said, Linux won't be replacing Solaaris on my machines. If
> Sun does force me to drop Solaris, I'll be going to BSD, not Linux.
> And if that means dropping Sun hardware as well, well that will be
> their own damn fault, won't it?
>
Oh I am sure it will still be called Solaris, and I suspect all the
software on top of the kernel will be the same as before. All that
will change is that they will rip out the current Solaris kernel
and bolt in a Linux one.
Lance
news:kkefma...@192.168.42.254...
> Oh I am sure it will still be called Solaris, and I suspect all the
> software on top of the kernel will be the same as before. All that
> will change is that they will rip out the current Solaris kernel
> and bolt in a Linux one.
>
And what would that get them? Nothing. They would spend too much time and
resources making sure that it worked with no benefit in return. Changing
the kernel does not mean that all the software would run on it.
Lance
Rich Teer
> Oh I am sure it will still be called Solaris, and I suspect all the
> software on top of the kernel will be the same as before. All that
> will change is that they will rip out the current Solaris kernel
> and bolt in a Linux one.
Oops, sorry. There goes your credibility. Sun may be doing some,
ummm, questionable things with this Linux stuff, but they're not
stupid enough to replace the Solaris kernel with the Linux one!
The Solaris kernel is lightyears ahead of the Linux one, no
matter how "cool" the latter is.
bit-b...@maney.org
: bit-b...@maney.org writes:
:> In comp.sys.sun.misc Jonathan Buzzard <jona...@buzzard.org.uk> wrote:
:>: In article <amd0ta$de7$1...@newsread.stdio.com>,
:>: bit-b...@maney.org writes:
:>
:> [...]
:>
:>:> No. SunOS is the operating system. Solaris is the operating environment.
:>:> Basically Solaris is SunOS (the operating system) + a slew of applications
:>:> that give a certain "look-n-feel" (the operating environment).
:>
:>: Not really the internals of the SunOS 4.1.x (aka Solaris 1.x) and
:>: Solaris 2.x onwards are entirely different. One (SunOS 4.1.x) is
:>: a BSD Unix and the other is a SysV Unix.
:>
:> It's still SunOS with the OE on top of it that is called Solaris.
:> Whether or not the Sun OS under it is BSD or SysV based is completely
:> irrelevant in this context.
: Not really, the point is that a kernel change was forced on you.
What? The original poster asked what the difference was between
SunOS and Solaris. As I said earlier, the difference is that
SunOS is the OS and Solaris is the OE. Whether or not the version
of SunOS is BSD or SysV derived doesn't matter.
No one "forced" anything on me. I could have stayed with the
BSD-derived SunOS 4.x of I could have switched to the SysV-derived
SunOS 5.x. Either way, it has no bearing on what the OP asked.
[...]
Barry Margolin
You could have said the same thing about the conversion from BSD-based
SunOS 4.x/Solaris 1.x to SysV-based SunOS 5.x/Solaris 2.x. Yes, there's
quite a bit of work required if you replace the underlying kernel, but the
benefit is that you can take advantage of resources that are available for
the new kernel. In the case of Solaris 2.x, they could collaborate with
AT&T and be more compatible with 3rd-party software that was written to the
SysV API. Now that Linux is becoming more popular, they may see it as an
attractive target; there are lots of Linux applications, it may be easier
for them to hire Linux developers, and they can take advantage of the
changes that are made to the Linux kernel by others.
I also suspect there's probably less work involved in porting their
applications to Linux these days. Linux is POSIX-conformant, isn't it?
When they converted from SunOS 4 to SunOS 5 they had to fix all of the
applications that made use of BSD-specific APIs. Most of the things that
will need to be changed will be tools that are kernel-specific (like "ps"
and "netstat") and some administrative tools. But in many of these cases
they can simply use the Linux versions of the commands (unless there are UI
incompatibilities and they want to retain the Solaris UI).
--
Barry Margolin, bar...@genuity.net
Genuity, Woburn, MA
*** DON'T SEND TECHNICAL QUESTIONS DIRECTLY TO ME, post them to newsgroups.
Please DON'T copy followups to me -- I'll assume it wasn't posted to the group.
Barry Margolin
>What? The original poster asked what the difference was between
>SunOS and Solaris. As I said earlier, the difference is that
>SunOS is the OS and Solaris is the OE. Whether or not the version
>of SunOS is BSD or SysV derived doesn't matter.
While that's the strict definition, it's also extremely common to use
"SunOS" to refer to Solaris 1.x, and "Solaris" to refer to Solaris 2.x.
This is because the name Solaris wasn't introduced until Solaris 2.x was
being deployed. Prior to that, the entire system was calls SunOS, and they
retroactively started calling the systems based on SunOS 4.1.x "Solaris
1.x". The renaming didn't really catch on with the users, who considered
the SunOS/Solaris distinction to be the version, not kernel vs. OE.
Paul Eggert
> I also suspect there's probably less work involved in porting their
> applications to Linux these days. Linux is POSIX-conformant, isn't it?
No. It is close, but it doesn't conform exactly, nor does it claim
to. However, the places where it fails to conform are small, and are
typically only a small part of any porting hassle.
Jonathan Buzzard
Rich Teer <ri...@rite-group.com> writes:
> On Fri, 20 Sep 2002, Jonathan Buzzard wrote:
>
>> Oh I am sure it will still be called Solaris, and I suspect all the
>> software on top of the kernel will be the same as before. All that
>> will change is that they will rip out the current Solaris kernel
>> and bolt in a Linux one.
>
> Oops, sorry. There goes your credibility. Sun may be doing some,
> ummm, questionable things with this Linux stuff, but they're not
> stupid enough to replace the Solaris kernel with the Linux one!
>
> The Solaris kernel is lightyears ahead of the Linux one, no
> matter how "cool" the latter is.
Not today for sure, but three, four, five years from now that
may well no longer be the case.
Drazen Kacar
> Barry Margolin <bar...@genuity.net> writes:
>
> > I also suspect there's probably less work involved in porting their
> > applications to Linux these days. Linux is POSIX-conformant, isn't it?
>
> No. It is close, but it doesn't conform exactly, nor does it claim to.
But Sun claims it's compliant. "Since both the Solaris OE and Linux are
POSIX 1003.1 compliant [...]"
http://soldc.sun.com/migration/linux/l2s_plan.html
--
.-. .-. I don't work here. I'm a consultant.
(_ \ / _)
| da...@willfork.com
|
I R A Darth Aggie
Cyphe...@nyc.rr.com <Cyphe...@nyc.rr.com>, in
<V0Ki9.10780$GJ3.2...@twister.nyc.rr.com> wrote:
+> In comp.unix.solaris Bryan Althaus <br...@panix.com.invalid> wrote:
+> >
+> > Rich,
+> >
+> > That's "kewl" not "cool". I am shocked at how bent out of shape people
+> > are getting at the fact a UNIX company like Sun has realized that it can
+> > make money selling Linux in edge area's where they couldn't sell Solaris.
+>
+> Sun strategy is to sell them as part of a larger package
+> involving SPARC servers.
Linux on the desktop, Solaris in the back office?
Maybe. I could see that as a way of assaulting Microsoft and either
keeping them out of the server room, or booting them out of the server
room. But I still am not sure if that's really Sun's strategy.
Steve Kappel
> In article <Pine.GSO.4.44.020920...@mars.rite-group.com>,
> Rich Teer <ri...@rite-group.com> writes:
>> On Fri, 20 Sep 2002, Jonathan Buzzard wrote:
>>
>>> Oh I am sure it will still be called Solaris, and I suspect all the
>>> software on top of the kernel will be the same as before. All that
>>> will change is that they will rip out the current Solaris kernel
>>> and bolt in a Linux one.
>>
>> Oops, sorry. There goes your credibility. Sun may be doing some,
>> ummm, questionable things with this Linux stuff, but they're not
>> stupid enough to replace the Solaris kernel with the Linux one!
>>
>> The Solaris kernel is lightyears ahead of the Linux one, no
>> matter how "cool" the latter is.
>
> Not today for sure, but three, four, five years from now that
> may well no longer be the case.
And pigs may fly...
This is the same kind of BS that a few years ago suggested Windows
would eliminate all other operating systems in 5 years. Am I
the only one who hears the Linux whine and thinks it sounds just
like the Windows whine?
Tim Bradshaw
> You could have said the same thing about the conversion from BSD-based
> SunOS 4.x/Solaris 1.x to SysV-based SunOS 5.x/Solaris 2.x. Yes, there's
> quite a bit of work required if you replace the underlying kernel, but the
> benefit is that you can take advantage of resources that are available for
> the new kernel. In the case of Solaris 2.x, they could collaborate with
> AT&T and be more compatible with 3rd-party software that was written to the
> SysV API. Now that Linux is becoming more popular, they may see it as an
> attractive target; there are lots of Linux applications, it may be easier
> for them to hire Linux developers, and they can take advantage of the
> changes that are made to the Linux kernel by others.
The problem, it seems to me, with doing this is it leaves Sun with
essentially nothing to offer. I don't know where their profits come
from but I bet it's mostly big-iron machines, and if they do all the
work to make Linux scale to these systems then everyone else can, and
will, benefit from that. But Sun won't significantly benefit from all
the linux stuff, because almost all of that is aimed at the kind of
machines that students can buy (PCs and PDAs) and the kind of things
that students want to do (play quake and surf the web). If they want
to benefit from all the stuff linux has to offer they need to start
trying to make a lot of money out of desktop boxes, and I don't think
there is a lot of profit in that market any more.
It also seems to me that this is the wrong way up: I don't really care
what the kernel does - I'm just a user. The thing that makes Linux
cool for me isn't that it has an ideologically sound kernel, it's that
it has a choice of decent guis that I don't have to compile myself,
and packet filtering that I don't have to install myself and so on.
If I was Sun I would do the following: keep the SunOS kernel, but
produce a very good linux compatibility layer around it so that linux
applications became *trivial* to port. In fact, other than possibly
having to be recompiled for sparc, linux applications should think
that they *were running on linux*. As far as I can see this requires
two things: a system call interface that looks like Linux's (or
possibly just a libc which looks like linux's), and enough of the
linux utilities that the system smells like linux to software. The
latter ought to be pretty simple compared to the former, since the
linux utilities just depend on the system call interface. On top of
this they could do some virtualization stuff and lo and behold you
have linux applications seamlessly hosted on top of SunOS, and
coexisting with Solaris applications.
This smells, to me, suspiciously like what IBM have done, except I
presume they have some completely non-unix OS sitting under everything
and doing the virtualization.
--tim
Thomas Dehn
"Tim Bradshaw" <t...@cley.com> wrote:
> It also seems to me that this is the wrong way up:
> I don't really care what the kernel does - I'm
> just a user. The thing that makes Linux cool
> for me isn't that it has an ideologically sound kernel, it's that
> it has a choice of decent guis that I don't have to compile myself,
From a Solaris/Sparc PoV, 'decent gui' requires that
it can flawlessly run existing applications created
for OpenWindows and CDE even
in 8bit (colormap flashing and friends)
and 8+24 bit visuals.
Thomas
Steven Shelikoff
>In article <slrnaon4a0....@gurcragntba.pbz>, sy_n...@gurcragntba.pbz (I R A Darth Aggie) writes:
>
>[14 lines snipped]
>
>>Linux on the desktop,
>
>Until a Linux desktop comes along that is *indistinguishable* from Windows, it
>ain't gonna happen.
It doesn't have to be indistinguishable from windows. All the various
flavors of windows are not indistinguishable from each other. The only
thing Linux needs to make inroads on the desktop that it currently
doesn't have is a good native office suite. And no, not StarOffice, at
least not 5.2. It has to be fully compatible with
Word/Excel/PowerPoint/Outlook.
Steve
Steven Shelikoff
>In article <3d8ba9b...@cnews.newsguy.com>, shel...@yawho.com (Steven Shelikoff) writes:
>>On 20 Sep 2002 21:55:43 GMT, hu...@nospam.huge.org.uk (Huge) wrote:
>>
>>>In article <slrnaon4a0....@gurcragntba.pbz>, sy_n...@gurcragntba.pbz (I R A Darth Aggie) writes:
>>>
>>>[14 lines snipped]
>>>
>>>>Linux on the desktop,
>>>
>>>Until a Linux desktop comes along that is *indistinguishable* from Windows, it
>>>ain't gonna happen.
>>
>>It doesn't have to be indistinguishable from windows. All the various
>>flavors of windows are not indistinguishable from each other. The only
>>thing Linux needs to make inroads on the desktop that it currently
>>doesn't have is a good native office suite.
>
>discussing whether this is a good thing or not, or the merits or
>otherwise of the Windows UI) unless they cannot tell the difference
>between what they have now and what they've been given.
>
>Of course, they may be forced into it, by their employers for example,
>but "Joe Sixpack" is not going to use Linux unless he cannot
>distinguish it from Windows. And that includes installing, configuring
>and using it.
lol. Well, then we'd all still be using Windows 1.0. Well, maybe not
1.0. But definitely 95.
It has to be as easy to install, configure and use as windows. But it
definitely doesn't have to be indistinguishable. Actually, it doesn't
even have to be easy to install as long as you can buy PC's with it
pre-installed, like windows. The vast majority of windows users have
never actually installed windows.
Stev
Rich Teer
> Not today for sure, but three, four, five years from now that
> may well no longer be the case.
That makes the somewhat dubious assumption that the Solaris
kernel will stand still while Linux plays catchup. In that
same 3, 4, or 5 years span, Solaris will also have improved.
Dan
"Huge" <hu...@nospam.huge.org.uk> wrote in message
news:amgbkh$iru$5...@anubis.demon.co.uk...
> >
> >>In article <slrnaon4a0....@gurcragntba.pbz>,
sy_n...@gurcragntba.pbz (I R A Darth Aggie) writes:
> >>
> >>[14 lines snipped]
> >>
> >>>Linux on the desktop,
> >>
> >>Until a Linux desktop comes along that is *indistinguishable* from
Windows, it
> >>ain't gonna happen.
> >
> >It doesn't have to be indistinguishable from windows. All the various
> >flavors of windows are not indistinguishable from each other. The only
> >thing Linux needs to make inroads on the desktop that it currently
> >doesn't have is a good native office suite.
>
> discussing whether this is a good thing or not, or the merits or
> otherwise of the Windows UI) unless they cannot tell the difference
> between what they have now and what they've been given.
>
> Of course, they may be forced into it, by their employers for example,
> but "Joe Sixpack" is not going to use Linux unless he cannot
> distinguish it from Windows. And that includes installing, configuring
> and using it.
>
> >least not 5.2. It has to be fully compatible with
> >Word/Excel/PowerPoint/Outlook.
>
>
Sorry, I totally disagree. This is the new republican age. Productivity is
key. You don't like your desktop, find another job. No more printers, no
more internet access, no more overtime. Christmas bonus is you get to come
back to work next year.
Scott McNealy made a good point. "white" box PC's have the life span of a
banana. SS5's are still running, lots of servers still running 2.6. How
many Windows 95 exchange servers? This could be the motivation to finally
get internet applications really working and integrated correctly and
finally down to browser only workstations. Client/server back with a
vengance.
Consumer desires are the last thing looked at when it comes to desktop sales
strategies. The sales are in the business world.
yalu
> Oops, sorry. There goes your credibility. Sun may be doing some, ummm,
> questionable things with this Linux stuff, but they're not stupid enough
> to replace the Solaris kernel with the Linux one!
>
> The Solaris kernel is lightyears ahead of the Linux one, no matter how
> "cool" the latter is.
Motivation please :-)
People often make this kind of suggestions without explaining what they
mean; I don't know anything about kernel internals so I have no clue at
what solaris or linux might be better.
--
mail: frankpuntvandammebijstudent-kuleuven-ac-belgie
homepage: www.student.kuleuven.ac.be/~m9917684
jabber: ya...@jabber.com
yalu
> Maybe. I could see that as a way of assaulting Microsoft and either
> keeping them out of the server room, or booting them out of the server
> room. But I still am not sure if that's really Sun's strategy.
Anyone with a head on their shoulder keeps windows out of the server room.
Windows is barely suited as a desktop os, let alone as a server.
yalu
> Until a Linux desktop comes along that is *indistinguishable* from
> Windows, it ain't gonna happen.
Nonsense. Just as if the windows desktop is more "userfriendly" or
"intuitive" or any other crap then a kde or gnome (dunno about CDE ;) ).
If people don't want to run a near-windows gnu/linux desktop, they're just
being irrational or conservative.
yalu
> Of course, they may be forced into it, by their employers for example,
> but "Joe Sixpack" is not going to use Linux unless he cannot
> distinguish it from Windows. And that includes installing, configuring
> and using it.
Way too much generalisation...
My sister does nearly everything under gnu/linux right now and I know of
dozens of other people who don't mind switching desktops. btw windowmaker
is the window manager of choice, so it doesn't even look like windows,
rather like NeXT (not that I ever saw THAT running).
yalu
> : Agreed. That too.
>
> I remeber when DR DOS 5.0 was being used over Microsoft DOS. So it's
> possible, but only when Microsoft sits still and others progress.
Reverse engineering sucks... btw ms doesn't really "progress", just
"change" (their formats and stuff). And even if they did that, you can
only try to imitate their stuf, it will never be perfect. btw I have yet
to meet an excel file I dannot open with either openoffice.org or
gnumeric, or a word file that doesn't fit OOo. Even kword does a good job
at that these days.
Joerg Schilling
Jonathan Buzzard <jona...@buzzard.org.uk> wrote:
>> The Solaris kernel is lightyears ahead of the Linux one, no
>> matter how "cool" the latter is.
>
>Not today for sure, but three, four, five years from now that
>may well no longer be the case.
I hear this statement from Linux followers for more than five years now....
--
EMail:jo...@schily.isdn.cs.tu-berlin.de (home) Jörg Schilling D-13353 Berlin
j...@cs.tu-berlin.de (uni) If you don't have iso-8859-1
schi...@fokus.gmd.de (work) chars I am J"org Schilling
URL: http://www.fokus.gmd.de/usr/schilling ftp://ftp.fokus.gmd.de/pub/unix
Joerg Schilling
Huge <hu...@nospam.huge.org.uk> wrote:
>In article <ey3znuc...@cley.com>, Tim Bradshaw <t...@cley.com> writes:
>
>[43 lines snipped]
>
>>This smells, to me, suspiciously like what IBM have done, except I
>>presume they have some completely non-unix OS sitting under everything
>>and doing the virtualization.
>
>recently.)
This would be
- Completely new
- Not help the potential users.
People who like to use Linux on big IBM machines use it because there are no more
hackers around for the old OS. So they run Linux on a virtual machine in the
same housing.
If Linux really runs on those machines: Linux does not support OS domains.
yalu
> Sorry, but you people obviously have no contact with non-IT type computer
> users. People for whom the PC is of no intrinsic interest - so far as
> they are concerned it's like the photocopier, a black box whose inner
> workings are a mystery of no interest. The people who peer over my shoulder
> at work (I have a Blade 100 on my desk) and say "What kind of PC is
> that?"
And you answer?
yalu
>>Way too much generalisation...
>>
>>My sister does nearly everything under gnu/linux right now and I know of
>>dozens of other people
>
> I support some 8000 users. I've never been asked for a Linux desktop.
> Ever.
Wasn't exactly what I meant. People these days who actually ask linux have
a strong motivation to get rid of windows and are usually quiet
technically interested.
What I meant is that if one day you put them behind a linux box and you
make sure they won't miss anything (eg run the same apps or similar
replacements) the majority won't complain. So the gui is somewhat
different? It doesn't have a taskbar but a window list, windows explorer
is suddenly called konqueror etc. etc. The users you support haven't
installed or configured their windows boxes and so they won't have to do
this with the linux boxes either.
yalu
>>Nonsense. Just as if the windows desktop is more "userfriendly" or
>>"intuitive" or any other crap then a kde or gnome (dunno about CDE ;) ).
>>If people don't want to run a near-windows gnu/linux desktop, they're
>>just being irrational or conservative.
>
> Eduhead drooling. Go talk to some real users who pay bills.
Well I guess it's better to be an eduhead then a 21-inch Solarishead.
Steven Shelikoff
>In article <3d8bb94a...@cnews.newsguy.com>, shel...@yawho.com (Steven Shelikoff) writes:
>>On 20 Sep 2002 23:40:01 GMT, hu...@nospam.huge.org.uk (Huge) wrote:
>
>[22 lines snipped]
>
>>>Of course, they may be forced into it, by their employers for example,
>>>but "Joe Sixpack" is not going to use Linux unless he cannot
>>>distinguish it from Windows. And that includes installing, configuring
>>>and using it.
>>
>>lol. Well, then we'd all still be using Windows 1.0. Well, maybe not
>>1.0. But definitely 95.
>>
>>It has to be as easy to install, configure and use as windows. But it
>>definitely doesn't have to be indistinguishable.
>
>Sorry, but you people obviously have no contact with non-IT type computer
>users. People for whom the PC is of no intrinsic interest - so far as
>they are concerned it's like the photocopier, a black box whose inner
>workings are a mystery of no interest. The people who peer over my shoulder
>at work (I have a Blade 100 on my desk) and say "What kind of PC is that?"
Well sure. And they would say the exact same thing if you were running
win95 and all they'd seen before was win3.11.
Steve
yalu
> In article <pan.2002.09.21.10...@mail.com>, yalu <frank...@mail.com> writes:
>>On Fri, 20 Sep 2002 23:21:04 +0200, I R A Darth Aggie wrote:
>>
>>> Maybe. I could see that as a way of assaulting Microsoft and either
>>> keeping them out of the server room, or booting them out of the server
>>> room. But I still am not sure if that's really Sun's strategy.
>>
>>Anyone with a head on their shoulder keeps windows out of the server room.
>
> It's pretty obvious you're a student. When the business says "I want this",
> where "this" is (for example) an SQL Server based application, then you
> do it. You don't say to your CEO, "Oh, no, I don't want that in 'my' (*)
> computer room, for religious reasons".
>
> (* It's not "your" computer room. It's the business's.)
I'm understand that (that being difficult for a student obviously).
And I'm not talking about the Church of Emacs. I'm talking about
"performance", "reliability", "manageability", "total cost of ownership",
in short business talk. If your boss asks you to set up an SQL backend for
app $foo it's pretty easy to choose between ms sql server and mysql or
postgresql or oracle or maybe something Sun offers on unix/linux.
People normally don't ask for products but for solutions. Unless off
course you allready locked yourself in to some proprietary solution
*before* the question which solution to choose even arises - in the above
example I suppose you mean "ms sql server" with "sql server"? Why should
your boss care about which backend his sql based app works? If you're
allready using an ms sql based product, it's not a matter of avoiding
windows - it is too late allready :-)
(about the "booting windows out of the server room" being Sun's strategy
or not: yes I'd rather be using solaris then windows but it's even better
you can decide for that yourself if you think you can benefit from
migrating that particular windows based app to the os of your choice; in
the case of your sql product: afaik sql stuff is easy to port, and it
requires access to the source code. Or am I going too religious now?)
yalu
> Well, this is of no relevance, but "It's not a PC". To which the response
> is either incomprehension (most PC end-users do not realise that there are any
> other kind of computer) or the question "Well, what is it then?" I respond
> "a Sun workstation", and *then* we move onto incomprehension.
<nitpicking>
though sun users seem to hate to call their box a pc, a second meaning of
"pc" is a small system used for rather light-weight applications locally
(surf, mail, play). So you got a sparc-based pc and the intels in the
server room aren't pc's :)
</nitpicking>
Jonathan Buzzard
hu...@nospam.huge.org.uk (Huge) writes:
[SNIP]
>
> It's pretty obvious you're a student. When the business says "I want this",
> where "this" is (for example) an SQL Server based application, then you
> do it. You don't say to your CEO, "Oh, no, I don't want that in 'my' (*)
> computer room, for religious reasons".
It is your role as the technical guy on the ground to advise that it is
not a good idea with reasons. If they then insist against the better
judgement of those that will be installing and running the system then
it is in your best interests to inform them that you will not take any
responsibility for the mess when it all falls apart, and do this in
writing in paper and circulate it about a bit and start applying for
new jobs.
> (* It's not "your" computer room. It's the business's.)
And it is not the CEO's either. It is your job as the technical guy with
all the knowledge to advise the CEO on what are the best solutions
for the business. It is his job to take it on board.
I have done it, and I have found that most bosses will listen to the
advice you give.
Jonathan Buzzard
>
>> Oops, sorry. There goes your credibility. Sun may be doing some, ummm,
>> questionable things with this Linux stuff, but they're not stupid enough
>> to replace the Solaris kernel with the Linux one!
>>
>> The Solaris kernel is lightyears ahead of the Linux one, no matter how
>> "cool" the latter is.
>
> Motivation please :-)
>
> People often make this kind of suggestions without explaining what they
> mean; I don't know anything about kernel internals so I have no clue at
> what solaris or linux might be better.
>
At the moment Solaris (or SunOS 5.x) is better if you have more than
2 CPU's, in the area of NFS, and realtime stuff as far as I am aware.
In some areas Linux is better than Solaris, most noticably LVM and
journeling filesystems. Primarly because this is a third part add
on for Solaris, and it is native with Linux.
Alex Buell
> >> Eduhead drooling. Go talk to some real users who pay bills.
> >
> >Well I guess it's better to be an eduhead then a 21-inch Solarishead.
>
> Don't worry. You'll learn. Once you've worked in IT for a few years.
Anyway, having a 21" is brilliant. I bet he's jealous.
--
I live in my own little world, but it's okay, everyone knows me there.
http://www.munted.org.uk (updated 10 September 2002)
Steven Shelikoff
>On Sat, 21 Sep 2002 14:45:36 +0200, Huge wrote:
>
>> Well, this is of no relevance, but "It's not a PC". To which the response
>> is either incomprehension (most PC end-users do not realise that there are any
>> other kind of computer) or the question "Well, what is it then?" I respond
>> "a Sun workstation", and *then* we move onto incomprehension.
>
><nitpicking>
>
>though sun users seem to hate to call their box a pc, a second meaning of
>"pc" is a small system used for rather light-weight applications locally
>(surf, mail, play). So you got a sparc-based pc and the intels in the
>server room aren't pc's :)
>
></nitpicking>
lol. I was going to say the same thing. It's not a wintel, but it is a
PC.
Steve
Chris Newport
>
> In article <amhl8r$kee$1...@news.cs.tu-berlin.de>, j...@cs.tu-berlin.de (Joerg Schilling) writes:
> From personal professional experience, I know this to be untrue. For
> example, you can run Linux under VM on a S/390 [Ooops, sorry,
> z/Server... :o( ] so you need VM people to set it up.
>
> The motivations to run Linux on Big Iron are varied and many, but "no more
> hackers" is none of them.
>
> >If Linux really runs on those machines:
>
> Which it does.
>
> > Linux does not support OS domains.
>
> Quite correct.
Linux will happily run within a domain on IBM or Sun hardware, but
that is not really the issue. Without luxadm and similar tools you
cannot manage the storage arrays that typically come with such boxes.
Linux has a good place at the edge of the data centre, and in a very
few specialist applications it has a place on the bigger stuff.
Until the storage and data security issues are resolved that is where
the story ends. To resolve these issues you need to get Veritas
involved.
Most people have missed the real point of Sun's new LX50 machine.
The *BIG* advance is that you can now buy an IA32 machine which
comes *as standard* with both Linux and Solaris, and BOTH ARE FULLY
SUPPORTED. Nobody else supplies a fully supported Linux solution.
IBM, for example, will tell you that DeadRat will run on their
systems, but if you call IBM for support they will tell you to
call DeadRat. At this point any responsible data centre manager
tells the salesdroid to take a walk and not return.
Corporate users need a fully supported solution, all the way from
having Sun Professional Services design and specify the entire
package, through to having a single point of contact for a support
agreement such as Sun's Platinum Support which WILL have a
properly qualified engineer on site within 2 hours and have the
problem fixed within 4 hours. Without an SLA the orders will not
get signed.
The Linux on Sparc issues are rather different. In most cases Linux
is appropriate on older Sun hardware which is at the end of
Solaris support but still has a usefull task to perform. Here the
user is expecting to provide his own support and fully understands
that when it breaks he is on his own. Sun do not (yet ?) offer
a supported Linux for Sparc, and there are no indications that
they intend to do so.
Sun's strategy is clear. They are now the market leader in the
data centre and need to push outwards to evict Microsoft based
systems. first from the edges just inside the data centre doors,
and then start making inroads onto the desktop. Sun have already
pushed HP and SGI to the margins on the Unix desktop, the
Windows based desktop is a tougher nut to crack.
Scott has hinted at a launch early next year, expect to see
something which will replace the typical office desktop with
a low cost product which offers compatibility as well as a
greatly reduced cost of ownership. Without some very significant
cost advantages the corporate buyers will not be impressed.
Dan
"Chris Newport" <c...@NOSPAM.netunix.com> wrote in message
news:3D8C861A...@NOSPAM.netunix.com...
He already did this week. Wednesday I think.
yalu
>>What I meant is that if one day you put them behind a linux box and you
>>make sure they won't miss anything (eg run the same apps or similar
>>replacements) the majority won't complain.
>
> Tee-hee. Your naivety is quite amusing. Users complain about *everything*.
OK, you win.
yalu
>> Don't worry. You'll learn. Once you've worked in IT for a few years.
>
> Anyway, having a 21" is brilliant. I bet he's jealous.
Indeed I am, since I re-measured my sun monitor and saw that it was only
20 inches.
yalu
> Don't worry. You'll learn. Once you've worked in IT for a few years.
Yes pa!
Alan Coopersmith
|In some areas Linux is better than Solaris, most noticably LVM and
|journeling filesystems. Primarly because this is a third part add
|on for Solaris, and it is native with Linux.
Solaris UFS has had a logging option since Solaris 7, and Solaris 8 and
later include LVM. Try again.
--
________________________________________________________________________
Alan Coopersmith al...@alum.calberkeley.org
http://soar.Berkeley.EDU/~alanc/ aka: Alan.Coo...@Sun.COM
Working for, but definitely not speaking for, Sun Microsystems, Inc.
Rich Teer
> People often make this kind of suggestions without explaining what they
> mean; I don't know anything about kernel internals so I have no clue at
> what solaris or linux might be better.
Solaris scales better, is more stable under high loads, and has
better support for domains and hot swap.
And its advocates don't tend to drool so much... :-)
Alex Buell
> On Sat, 21 Sep 2002 16:37:56 +0200, Alex Buell wrote:
>
> >> Don't worry. You'll learn. Once you've worked in IT for a few years.
> >
> > Anyway, having a 21" is brilliant. I bet he's jealous.
>
> Indeed I am, since I re-measured my sun monitor and saw that it was only
> 20 inches.
Like everyone says, having an extra inch really does help.
Rich Teer
> Anyone with a head on their shoulder keeps windows out of the server room.
True. Unfortuantely, this sort of dicision isn't always
made by qualified people. More often than not, managemnt
gets involved, and fscks it up.
> Windows is barely suited as a desktop os, let alone as a server.
Agreed.
Rich Teer
> If people don't want to run a near-windows gnu/linux desktop, they're just
> being irrational or conservative.
I disagree, to an extent. GNOME is a near-windoze Linux desktop,
and right now, I don't like it. I don't want a fscking stupid
"task bar". If I minimise an application, I want it on the desktop.
I've not played with GNOME 2.0 Beta 2 yet, but Beta 1 didn't do much
for me.
tuca...@whodis.org
: where "this" is (for example) an SQL Server based application, then you
: do it. You don't say to your CEO, "Oh, no, I don't want that in 'my' (*)
: computer room, for religious reasons".
Hm. Actually, I do do this, and it works. I've thrown away more PC-based
servers in the past year than I can count. Whenever I am asked the above
question, my response is "do you want it to be reliable" and, when they
say yes, "then throw the PC away."
Religion isn't a part of it. More like common sense.
tuca...@whodis.org
: journeling filesystems. Primarly because this is a third part add
WHAT?!
Chris Newport
>
> In article <3D8C861A...@NOSPAM.netunix.com>, Chris Newport <c...@NOSPAM.netunix.com> writes:
>
> [35 lines snipped]
>
> >IBM, for example, will tell you that DeadRat will run on their
> >systems, but if you call IBM for support they will tell you to
> >call DeadRat.
>
> Support for IBM and the CEO of Red Hat, both of whom I have spoken to
> about this.
>
> [4 lines snipped]
>
> >Corporate users need a fully supported solution, all the way from
> >having Sun Professional Services design and specify the entire
> >package, through to having a single point of contact for a support
> >agreement such as Sun's Platinum Support which WILL have a
> >properly qualified engineer on site within 2 hours and have the
> >problem fixed within 4 hours. Without an SLA the orders will not
> >get signed.
>
Either I pay too much attention to Scott's presentations or IBM
are offering mixed messages. I suspect that IBM need to make the
the public commitment as global policy rather than making a local
arrangement for Europe.
Thomas Dehn
"Rich Teer" <ri...@rite-group.com> wrote:
> On Sat, 21 Sep 2002, yalu wrote:
>
> > If people don't want to run a
> > near-windows gnu/linux desktop, they're just
> > being irrational or conservative.
>
> I disagree, to an extent. GNOME is a near-windoze Linux desktop,
> and right now, I don't like it. I don't want a fscking stupid
> "task bar". If I minimise an application, I want it on the desktop.
I vote "task bar" for most poorly designed
basic GUI functionality ever.
Thomas
Thomas Dehn
"Huge" <hu...@nospam.huge.org.uk> wrote:
> In article <ey3znuc...@cley.com>, Tim Bradshaw <t...@cley.com> writes:
>
> [43 lines snipped]
>
> >This smells, to me, suspiciously like what IBM have done, except I
> >presume they have some completely non-unix OS sitting under everything
> >and doing the virtualization.
>
> AFAIK, Linux runs natively on all the IBM iron
> (the AS/400 only recently.)
I wouldn't call anything which is running inside
of a VM "running natively".
Thomas
Wim Bakker
Rich Teer <ri...@rite-group.com> wrote:
> On Sat, 21 Sep 2002, yalu wrote:
>
> > People often make this kind of suggestions without explaining what they
> > mean; I don't know anything about kernel internals so I have no clue at
> > what solaris or linux might be better.
>
> Solaris scales better, is more stable under high loads, and has
> better support for domains and hot swap.
>
> And its advocates don't tend to drool so much... :-)
>
On the other hand, linux has much better support and faster fixes .
I have a sun box at home, still waiting for a patch for driver
related problems with (a number of) scsi cd-rom's and cd-rw's.
Would it have been linux , I could have e-mailed the writer of the driver
and come to a solution in a short time, now I have to wait till sun
sees it fit (if ever) to come with a solution.
wb
Thomas Dehn
Do I understand correctly that you want untested
patches which might possibly fix your issue,
but break essential other functionality with a
high probability?
Thomas
yalu
>> I disagree, to an extent. GNOME is a near-windoze Linux desktop,
>> and right now, I don't like it. I don't want a fscking stupid
>> "task bar". If I minimise an application, I want it on the desktop.
>
> I vote "task bar" for most poorly designed
> basic GUI functionality ever.
There's better... imho the gnome taskbar is the one-eyed in the land of
blind then (my current taskbar has a start button, 1 icon and a clock. no
tasklist). it sits in the corner of the screen and takes maybe 8 cm. of
place.)
P.S. I vote for desktop icons. There is *always* a window over them.
yalu
> Hm. Actually, I do do this, and it works. I've thrown away more PC-based
> servers in the past year than I can count. Whenever I am asked the above
> question, my response is "do you want it to be reliable" and, when they
> say yes, "then throw the PC away."
Makes me wonder... are there any "real" data on reliability of intel/amd
based stuff versus sun based stuff? The lifespan of most computers are
10-15 years, mostly it are the disks who break first (if the memory doesn't
become faulty first), and you can put any brand of disks in a pc or sparc.
> Religion isn't a part of it. More like common sense.
--
Alan Coopersmith
|On the other hand, linux has much better support and faster fixes .
For users running boxes at home without support contracts.
|I have a sun box at home, still waiting for a patch for driver
|related problems with (a number of) scsi cd-rom's and cd-rw's.
|Would it have been linux , I could have e-mailed the writer of the driver
|and come to a solution in a short time, now I have to wait till sun
|sees it fit (if ever) to come with a solution.
Unless the Linux driver writer no longer feels like maintaining it, got
hit by a bus, changed e-mail addresses and didn't tell anyone, doesn't
have the exact same hardware configuration as you, or for some other
reason can't/won't fix your bug.
With Sun, you can be sure someone will be there to fix the bug, and if
you're willing to pay for support, can get someone on the phone 24 x 7
and can get bugs fixed relatively quickly. (It helps alot to know your
way around Sun tech support - if you've got a support contract and need
a bug fixed, you don't just call in and report it, you also have to add
the magic words "and I need a patch soon, so this bug needs to be
escalated".)
Jonathan Buzzard
hu...@nospam.huge.org.uk (Huge) writes:
> In article <4trhma...@192.168.42.254>, jona...@buzzard.org.uk (Jonathan Buzzard) writes:
>>In article <amhm0d$1aa$1...@anubis.demon.co.uk>,
>> hu...@nospam.huge.org.uk (Huge) writes:
>>
>>[SNIP]
>>>
>>> It's pretty obvious you're a student. When the business says "I want this",
>>> where "this" is (for example) an SQL Server based application, then you
>>> do it. You don't say to your CEO, "Oh, no, I don't want that in 'my' (*)
>>> computer room, for religious reasons".
>>
>>It is your role as the technical guy on the ground to advise that it is
>>not a good idea with reasons. If they then insist against the better
>>judgement of those that will be installing and running the system then
>>it is in your best interests to inform them that you will not take any
>>responsibility for the mess when it all falls apart, and do this in
>>writing in paper and circulate it about a bit and start applying for
>>new jobs.
>
>
Frankly if you can't persuade them to the soundness of your technical
arguments the problem is your abilities. I am sure your technical
ability to run the computers is more than adequate. However there
is more to life than this.
Jonathan Buzzard
yalu
>>Would it have been linux , I could have e-mailed the writer of the driver
>
> Assuming you could find him.
>
>>and come to a solution in a short time,
>
> Assuming he was still interested and prepared to talk to you.
In extremis, you can (hire someone to) fix it (yourself)
I R A Darth Aggie
Huge <hu...@nospam.huge.org.uk>, in
<amhj5b$1aa$2...@anubis.demon.co.uk> wrote:
+> Sorry, but you people obviously have no contact with non-IT type computer
+> users.
For the purposes of this thread, the home user and the small office
user is immaterial. We're talking corporate desktops and corporate
users, where a simple memo from the CEO to the CIO/CTO would make it
happen.
James
--
Consulting Minister for Consultants, DNRC
I can please only one person per day. Today is not your day. Tomorrow
isn't looking good, either.
I am BOFH. Resistance is futile. Your network will be assimilated.
I R A Darth Aggie
Huge <hu...@nospam.huge.org.uk>, in
<aminek$8r1$1...@anubis.demon.co.uk> wrote:
+> Bigotted nonsense. Computers are tools. If the best tool for the job is
+> a Windows box, then game over.
Name one.
Paul Eggert
> A PC doesn't last 10 to 15 years. Most people use windows and it
> wouldn't have any drivers for anything 10 to 15 years old. I doubt
> linsux will be any better.
If you look at the latest Netcraft host uptime survey, the best
GNU/Linux box (a Sun Cobalt box, by the way) is only #37 on the list.
Solaris doesn't even crack the top 50. So, by this (admittedly
limited) measure, GNU/Linux is quite a bit better than Solaris.
http://uptime.netcraft.com/up/today/top.avg.html
Perhaps a more realistic measure is the hosting network uptime survey.
Here, GNU/Linux has 5 of the top 10 netblocks. Solaris has 3.
http://uptime.netcraft.com/up/today/isp.avg.html
Now, we're just talking web servers here, which is perhaps not
Solaris's strongest suit these days. But these surveys match our own
company's experience: GNU/Linux is quite respectable compared to
Solaris in reliability, if it's managed well, and if you know its
limitations.
Steven Shelikoff
>Lance <nos...@nospam.org> writes:
>
>> A PC doesn't last 10 to 15 years. Most people use windows and it
>> wouldn't have any drivers for anything 10 to 15 years old. I doubt
>> linsux will be any better.
>
>If you look at the latest Netcraft host uptime survey, the best
>GNU/Linux box (a Sun Cobalt box, by the way) is only #37 on the list.
>Solaris doesn't even crack the top 50. So, by this (admittedly
>limited) measure, GNU/Linux is quite a bit better than Solaris.
>
>http://uptime.netcraft.com/up/today/top.avg.html
If you bothered to read the site FAQs, you'd see why neither Linux nor
Solaris could ever be anywhere near the top of the list even if they
really do have the longest uptimes. The fact that Linux turned up at 37
is an anomoly.
>Perhaps a more realistic measure is the hosting network uptime survey.
>Here, GNU/Linux has 5 of the top 10 netblocks. Solaris has 3.
>
>http://uptime.netcraft.com/up/today/isp.avg.html
Still not realistic. But the fact that Linux or Solaris is anywhere
near the top in this list is a remarkable achievement considering their
data collection methods.
>Now, we're just talking web servers here, which is perhaps not
>Solaris's strongest suit these days. But these surveys match our own
>company's experience: GNU/Linux is quite respectable compared to
>Solaris in reliability, if it's managed well, and if you know its
>limitations.
That's true. Although the stats at that site won't show that.
Steve
Steven Shelikoff
wrote:
>On 21 Sep 2002 18:43:21 -0700, Paul Eggert <egg...@twinsun.com> wrote:
>
>>Lance <nos...@nospam.org> writes:
>>
>>> A PC doesn't last 10 to 15 years. Most people use windows and it
>>> wouldn't have any drivers for anything 10 to 15 years old. I doubt
>>> linsux will be any better.
>>
>>If you look at the latest Netcraft host uptime survey, the best
>>GNU/Linux box (a Sun Cobalt box, by the way) is only #37 on the list.
>>Solaris doesn't even crack the top 50. So, by this (admittedly
>>limited) measure, GNU/Linux is quite a bit better than Solaris.
>>
>>http://uptime.netcraft.com/up/today/top.avg.html
>
>If you bothered to read the site FAQs, you'd see why neither Linux nor
>Solaris could ever be anywhere near the top of the list even if they
>really do have the longest uptimes. The fact that Linux turned up at 37
>is an anomoly.
In additional: The anomoly is likely that the system in position #37
recently switched to Linux (like, maybe 3 days ago since as of today,
the latest uptime is 3 days) but the numbers for average and max were
from the previous OS, which was not HP-UX, Linux, Solaris.
Steve
Chris Cox
> jona...@buzzard.org.uk (Jonathan Buzzard) writes in comp.unix.solaris:
> |In some areas Linux is better than Solaris, most noticably LVM and
> |journeling filesystems. Primarly because this is a third part add
> |on for Solaris, and it is native with Linux.
>
> Solaris UFS has had a logging option since Solaris 7, and Solaris 8 and
> later include LVM. Try again.
>
Yes, ufs has journaling and I'm guessing that you are referring
to DiskSuite as LVM, and while it is nice to see the addition of
named partitions, DiskSuite is still a far cry from the likes
of Veritas or the LVM in Linux. AFAIK, you still cannot resize
a ufs filesystem without unmounting it. I guess it's all
how you want to define things... I can certainly see the
poster's point.... in comparison, it's hard to conceive that
Sun's solution is comparable.
Paul Eggert
> That is why in general a PC cannot last 10 years. Sure you can have
> someone that has a 286 machine and a 10MB HD.
That is far older than 10 years. It's closer to 20. We're talking
maybe 1985.
> What can they do with it? Well, not much.
The same can be said for any circa-1985 box made by Sun. You can't
upgrade it to the latest version of Solaris. You can't run a modern
browser on it. You can run whatever apps you have stored on it, but
that's about it. Just like the ancient PC you mentioned.
Even a 10-year old Sun box is pretty old. Yes, if it's sitting in a
corner doing its thing, I suppose there's no point shutting it off.
But Sun won't support it and you would be wasting your time and money
actually developing or modifying it.
Don't get me wrong: I think Solaris SPARC is stabler than anything on
x86. But let's not go overboard about this. 10 years ago, I was
running Solaris 2.0 (or maybe 2.1; can't remember) on my home
SPARCstation ELC. I'd be crazy if I were still using that
hardware/software combination.
Stefaan A Eeckels
yalu <frank...@mail.com> wrote:
> People normally don't ask for products but for solutions.
Try again. The European Commission specifically asked for
an email solution based on Exchange and Outlook. Now they
have 80+ Exchange servers (I'm told at least one is
rebooting at any particular moment :-) and mail is regularly
lost.
One thing I've learned in my 25 years in the business is
that people ask for specific products, and want to interfere
at a very technical level. I still remember one of my early
customers back in 1980, who wanted an accounting package
written in COBOL, because "that is the business language".
We sold him an (expensive) copy of Cromemco COBOL and
installed the accounting system I had written in a thingy
called "Structured Basic". He was happy as a pig in shit ;-)
--
Stefaan
--
"One man alone can be pretty dumb sometimes, but for real bona fide
stupidity there ain't nothing can beat teamwork." -- Mark Twain
Stefaan A Eeckels
Cyphe...@nyc.rr.com wrote:
> > There was a claim in the late 1970s and early 1980s that Unix had
> > killed
> > operating systems research because no one would try anything else.
> > At the time, I didn't believe it. Today, I grudgingly accept that
> > the claim may be true (Microsoft notwithstanding).
>
> Oh, except for Microsoft? ;-)
In that they _did_ write an OS that's not a Unix clone.
But as it is a VMS clone, it doesn't break new ground
either.
> There are lots of proprietary OSes and kernels, and that involves a lot
> of R&D.
But they have not progressed beyond the Unix paradigm.
One of the craziest things is that we're creating OO
languages hand over fist, and run them in program/data
oriented OSes. What's the use of OO languages if the resulting
program is indistinguishable from one written in assembler?
Where's the OO environment to complement the OO language?
> Perhaps he was talking about commercial success?
Companies like MS depend on source code they no longer
know. I've worked for enough companies to know that
code is handed down to people who have neither the time
nor the competence to understand it. Those who wrote it
move on, and six months in a new job, they no longer
know why they wrote what they wrote.
MS cannot afford launching a new paradigm, and customers
aren't interested in a new paradigm. From now on, the best
we can hope for is new features, and another look. Just
like the car business. No innovation, just sell a new model
every four years. More gadgets, but it still only goes 55mph
on the highway.
It's no longer fun working in IT.
Thomas Dehn
"Bryan Althaus" <br...@panix.com.invalid> wrote:
> In comp.unix.solaris Thomas Dehn <thomas...@arcor.de> wrote:
I don't know, I haven't seen a Mac for many years.
Thomas
Thomas Dehn
"yalu" <frank...@mail.com> wrote:
> On Sat, 21 Sep 2002 21:16:48 +0200, Thomas Dehn wrote:
>
> >> I disagree, to an extent. GNOME is a near-windoze Linux desktop,
> >> and right now, I don't like it. I don't want a fscking stupid
> >> "task bar". If I minimise an application, I want it on the desktop.
> >
> > I vote "task bar" for most poorly designed
> > basic GUI functionality ever.
>
> There's better... imho the gnome taskbar is the one-eyed in the land of
> blind then (my current taskbar has a start button, 1 icon and a clock. no
> tasklist). it sits in the corner of the screen and takes maybe 8 cm. of
> place.)
And what happens if you minimize 20 application
windows? 50?
> P.S. I vote for desktop icons. There is *always* a window over them.
I don't have any windows over my 40 minimized application
windows. Most other windows are minimized, too :-)).
Thomas
yalu
>> People normally don't ask for products but for solutions.
>
> Try again.
Dunno about cobol (what is cobol? ; ) ) but your first example seems to
tell me that those people are generally WRONG or STUPID.
yalu
>> There's better... imho the gnome taskbar is the one-eyed in the land of
>> blind then (my current taskbar has a start button, 1 icon and a clock.
>> no tasklist). it sits in the corner of the screen and takes maybe 8 cm.
>> of place.)
>
> And what happens if you minimize 20 application windows? 50?
In the case of windows, you'll have to dual-head and put the taskbar on
the second head.
In my case: minimized or not, windows neatly appear in the Enlightenment
window list. A resizeable iconbox is also handy. Not to mention the pager.
Ah, I love E ;)
>> P.S. I vote for desktop icons. There is *always* a window over them.
>
> I don't have any windows over my 40 minimized application windows. Most
> other windows are minimized, too :-)).
--
Jonathan Buzzard
hu...@nospam.huge.org.uk (Huge) writes:
[SNIP]
>
> Neither would I.
>
> I repeat, Linux runs natively on all the IBM iron.
>
But on those machines with VM's there is little to be gained from
running natively, and in general on these machines the other operating
systems all run under VM's as well in normal usage.
Jonathan Buzzard
hu...@nospam.huge.org.uk (Huge) writes:
[SNIP]
>>Frankly if you can't persuade them to the soundness of your technical
>>arguments the problem is your abilities.
>
> You have no idea what you're talking about. And you're an arrogant sod,
> aren't you? You have no idea who I am or what I do, and yet you feel
> you can pronounce on my abilities.
If you cannot persuade none technical people that the techincal solution
they are specifically asking for is the wrong choice, then like it or
not there is a gap in your 'management' skills.
>>I am sure your technical
>>ability to run the computers is more than adequate. However there
>>is more to life than this.
>
> Indeed there is. And I hire and fire little tossers like you who think
> they know it all (fire, mostly) all the time. Grow up.
>
Yet you can't persuade people that they are making the wrong choices...
You know nothing about me either, personally I have always been able
to persuade those that mattered to take different solutions when it
mattered.
Joerg Schilling
Wim Bakker <w.a.b...@Xplanet.nl> wrote:
>> Solaris scales better, is more stable under high loads, and has
>> better support for domains and hot swap.
>>
>> And its advocates don't tend to drool so much... :-)
>>
>On the other hand, linux has much better support and faster fixes .
>I have a sun box at home, still waiting for a patch for driver
>related problems with (a number of) scsi cd-rom's and cd-rw's.
>Would it have been linux , I could have e-mailed the writer of the driver
>and come to a solution in a short time, now I have to wait till sun
>sees it fit (if ever) to come with a solution.
Do you like to stay sleeping and have sweet Linux dreams?
Cdrecord is developped on Solaris not on Linux.
--
EMail:jo...@schily.isdn.cs.tu-berlin.de (home) Jörg Schilling D-13353 Berlin
j...@cs.tu-berlin.de (uni) If you don't have iso-8859-1
schi...@fokus.gmd.de (work) chars I am J"org Schilling
URL: http://www.fokus.gmd.de/usr/schilling ftp://ftp.fokus.gmd.de/pub/unix
Wim Bakker
j...@cs.tu-berlin.de (Joerg Schilling) wrote:
> In article <20020921212903.40...@Xplanet.nl>,
> Wim Bakker <w.a.b...@Xplanet.nl> wrote:
>
> >> Solaris scales better, is more stable under high loads, and has
> >> better support for domains and hot swap.
> >>
> >> And its advocates don't tend to drool so much... :-)
> >>
> >On the other hand, linux has much better support and faster fixes .
> >I have a sun box at home, still waiting for a patch for driver
> >related problems with (a number of) scsi cd-rom's and cd-rw's.
> >Would it have been linux , I could have e-mailed the writer of the driver
> >and come to a solution in a short time, now I have to wait till sun
> >sees it fit (if ever) to come with a solution.
>
> Do you like to stay sleeping and have sweet Linux dreams?
>
> Cdrecord is developped on Solaris not on Linux.
what has cdrecord got to do with a boot up problem
with a number of scsi cdrom/cdrw drives on solaris9??
wb
Wim Bakker
al...@CSUA.Berkeley.EDU (Alan Coopersmith) wrote:
> Wim Bakker <w.a.b...@Xplanet.nl> writes in comp.unix.solaris:
> |On the other hand, linux has much better support and faster fixes .
>
> For users running boxes at home without support contracts.
>
Exactly, sun is good for business , not for home users,
if I go to the sun site for patches , I get an access denied
for patches because I don't have a spectrum support contract,
now how in the world would a home user have such a contract??
> |I have a sun box at home, still waiting for a patch for driver
> |related problems with (a number of) scsi cd-rom's and cd-rw's.
> |Would it have been linux , I could have e-mailed the writer of the driver
> |and come to a solution in a short time, now I have to wait till sun
> |sees it fit (if ever) to come with a solution.
>
> Unless the Linux driver writer no longer feels like maintaining it, got
> hit by a bus, changed e-mail addresses and didn't tell anyone, doesn't
> have the exact same hardware configuration as you, or for some other
> reason can't/won't fix your bug.
>
You would still have the source code and try fix it yourself
> With Sun, you can be sure someone will be there to fix the bug, and if
> you're willing to pay for support, can get someone on the phone 24 x 7
> and can get bugs fixed relatively quickly. (It helps alot to know your
> way around Sun tech support - if you've got a support contract and need
> a bug fixed, you don't just call in and report it, you also have to add
> the magic words "and I need a patch soon, so this bug needs to be
> escalated".)
Yes, but in what timeframe??
wb
Wim Bakker
hu...@nospam.huge.org.uk (Huge) wrote:
> In article <20020921212903.40...@Xplanet.nl>, Wim Bakker <w.a.b...@Xplanet.nl> writes:
>
> [17 lines snipped]
>
> >Would it have been linux , I could have e-mailed the writer of the driver
>
> Assuming you could find him.
>
never been a problem
> >and come to a solution in a short time,
>
> Assuming he was still interested and prepared to talk to you.
how often talks sun with a home user without support contract
about driver problems??
wb
>
>
>
>
Wim Bakker
"Thomas Dehn" <thomas...@arcor.de> wrote:
>
> "Wim Bakker" <w.a.b...@Xplanet.nl> wrote:
> > On Sat, 21 Sep 2002 17:22:04 GMT
> > Rich Teer <ri...@rite-group.com> wrote:
> > > Solaris scales better, is more stable under high loads, and has
> > > better support for domains and hot swap.
> > >
> > > And its advocates don't tend to drool so much... :-)
> > >
> > On the other hand, linux has much better support and faster fixes .
> > I have a sun box at home, still waiting for a patch for driver
> > related problems with (a number of) scsi cd-rom's and cd-rw's.
> > Would it have been linux , I could have e-mailed the writer of the driver
> > and come to a solution in a short time, now I have to wait till sun
> > sees it fit (if ever) to come with a solution.
>
> Do I understand correctly that you want untested
> patches which might possibly fix your issue,
> but break essential other functionality with a
> high probability?
>
Huhh?? what are you talking about? If my solaris9 has a
problem during boot time with quite a number of cdrom/cdrw drives
I expect a fix within a reasonable time from sun, available for
download to any user who has bought the sol9 oe (like I did).
wb
>
> Thomas
>
Stefaan A Eeckels
yalu <frank...@mail.com> wrote:
> On Sat, 21 Sep 2002 23:02:39 +0200, Stefaan A Eeckels wrote:
>
> >> People normally don't ask for products but for solutions.
> >
> > Try again.
>
> Dunno about cobol (what is cobol? ; ) ) but your first example seems to
> tell me that those people are generally WRONG or STUPID.
Those young ones. COBOL is "COmmon Business Oriented Language",
the first attempt at "bringing programming to the masses". In the
halcyon days of IBM, Sperry, Burroughs, ICL and other mainframes,
it was the language used for business applications. It's probably
still one of the most used languages in large businesses that
started using computers before 1990. As recently as Y2K, it got
a lot of press, so you must really a very young one ;-)
Also, don't forget that you can be WRONG, STUPID, and the person in
charge. In addition, managerial types have a very strong incentive
to believe Microsoft propaganda - isn't Microsoft the most successful
business of the moment? And who would _you_ believe? The guy with
the Microsoft business card, or the young one working for you?
Take care,
Stefaan A Eeckels
Lance <nos...@nospam.org> wrote:
> A CEO is not going to ask for an SQL server. He may say he wants "this"
> done. He could careless how it works, just that it works. The IT
> department is the one that decides how it will be implemented.
I've had CEOs issue diktats stating that henceforth, no
Unix machines could be bought, and every new development
would have to be on Windows.
Maybe you meant: A _competent_ CEO? Not many of those
around, today.
yalu
> Those young ones. COBOL is "COmmon Business Oriented Language",
> the first attempt at "bringing programming to the masses". In the
> halcyon days of IBM, Sperry, Burroughs, ICL and other mainframes,
> it was the language used for business applications. It's probably
> still one of the most used languages in large businesses that
> started using computers before 1990. As recently as Y2K, it got
> a lot of press, so you must really a very young one ;-)
20 fyi :)
I got my own computer since 1996 (windows 95 you know) and faintly got to
know what unix was a good 2 years ago (seeking a replacement for that
&%č§ā*^ų/ windows 95 and then downloaded Mandrake 7.0, which was, they
said, unix-alike). I didn't catch up too bad I guess.
> Also, don't forget that you can be WRONG, STUPID, and the person in
> charge. In addition, managerial types have a very strong incentive
> to believe Microsoft propaganda - isn't Microsoft the most successful
What I often ask myself is why people who know the best how economy works,
have studied, work in marketing or management, are such an easy victim for
m$ fud.
> business of the moment? And who would _you_ believe? The guy with
> the Microsoft business card, or the young one working for you?
Te numbers, I suppose.
> Take care,
I will ;)