Difference between solaris,dec,hp/ux etc
Jeff Turner
I am confused by the plethora of unix types out there. I mean what is the
point of having so many flavors of UNIX? What purpose do they serve? Can
somebody talk about the differences between sun solaris, BSD, HP/UX,
DEC/alpha and linux etc. I mean I know that they belong to different
companies and all. But where is the actual difference? What can solaris do
that HP/UX can't? Similarly what is the difference between Linux and
DEC/Alpha or BSD etc? Is there a web page that has a summary of the
similarities and differences between these types?
Thanks much,
Jeff
hac
their own.
When you are competing with another company, you can win with a better
price, or better service, or with a better product. In order for a
product to be "better", it must be different. Your idea of better may
be different than someone else's. Most often, these companies have
jealously guarded any changes that they have made, to use those changes
to win sales. Sometimes, they wake up for a moment, and realize that
the customers are getting pissed off by these differences, and the
companies get together and agree on some standard. But things designed
by committees tend to be pretty ugly, especially when every company has
to have someone on the committee, and every company has to get their
feature into the standard. Look at Motif.
Free or open source software, in contrast, is largely developed by the
people who want to use the software, not people trying to sell the
software. Not all users are developers, but the developers are most
often end users. The users have no reason to compete with each other,
and every reason to cooperate. So while proprietary UNIX flavors are
subject to forces that cause them to diverge, the open source
equivalents are subject to forces that cause them to converge. There
may be many Linux distributions and BSD flavors, but they have more in
common than the proprietary UNIX variants. Consensus is a powerful
force. Read the Makefiles for widely used programs, and you will
usually see ugly work-arounds for stupid differences between proprietary
Unices, but no distinction will be made between Linux distributions.
You could spend a life discussing all of the differences, and why they
came about. It took quite a few people many years to create them.
--
Howard Christeller Irvine, CA hchris...@home.com
Philip Brown
>force. Read the Makefiles for widely used programs, and you will
>usually see ugly work-arounds for stupid differences between proprietary
>Unices, but no distinction will be made between Linux distributions.
>
oh please. Let's not get too carried away with linux evangelism :-)
--
[Trim the no-bots from my address to reply to me by email!]
[ Do NOT email-CC me on posts. Pick one or the other.]
http://thomas.loc.gov/cgi-bin/bdquery/z?d105:SN01618:@@@D
The word of the day is mispergitude
Casper H.S. Dik - Network Security Engineer
hac <hchris...@home.com> writes:
>force. Read the Makefiles for widely used programs, and you will
>usually see ugly work-arounds for stupid differences between proprietary
>Unices, but no distinction will be made between Linux distributions.
Read commonly available source and wonder why it needs to be changed so
much to work on Linux and watch in amazement how programs ported to
Linux suddenly no longer compiler on the gazillion operating systems they
used to run on.
It's wrong to assume that all commercial vendors started with the
same Unix and pissed on it; and when it comes to Unix portability,
Linux does need some major include file reorganization. (Perhaps it
could try and follow the UNIX98 standard, even if it doesn't want
branding)
Casper
--
Expressed in this posting are my opinions. They are in no way related
to opinions held by my employer, Sun Microsystems.
Statements on Sun products included here are not gospel and may
be fiction rather than truth.
Charles Combs
> Hi,
>
> I am confused by the plethora of unix types out there. I mean what is the
> point of having so many flavors of UNIX? What purpose do they serve? Can
> somebody talk about the differences between sun solaris, BSD, HP/UX,
> DEC/alpha and linux etc. I mean I know that they belong to different
> companies and all. But where is the actual difference? What can solaris do
> that HP/UX can't? Similarly what is the difference between Linux and
> DEC/Alpha or BSD etc? Is there a web page that has a summary of the
> similarities and differences between these types?
>
>
Just in case you're interested, HP has a book out called "The HP-UX/Sun
Interoperability Cookbook". This book doesn't go into the differences between
the two OS's, rather, it provides a starting place for those sysadmins who have
to begin supporting the other OS.
Fairly well written and provides alot of info concerning commands,
configurations, etc.
--
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Charles "Rick" Combs | cco...@cadence.com
Senior Systems Engineer | Tel - (978)262-6251
Cadence Design Corp. | Fax - (978)262-6201
270 Billerica Rd. |
Chelmsford, MA 01824 |
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
"Ignorance of one's ignorance is the malady of the ignorant." O-
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
hac
> >usually see ugly work-arounds for stupid differences between proprietary
> >Unices, but no distinction will be made between Linux distributions.
> >
>
>
OK, I went too far with the make example, as you and several others
point out. But I stand by my point that pressure to differentiate
commercial offerings has not always served the UNIX user world well.
It's better now than during the 80's, but there's still room for
improvement. Why do so many people install the GNU tools? Could it be
that they want a more consistent environment across platforms? Could it
be that there are weaknesses and gaps in the commercial offerings?
Philip Brown
>...
>OK, I went too far with the make example, as you and several others
>point out. But I stand by my point that pressure to differentiate
>commercial offerings has not always served the UNIX user world well.
>It's better now than during the 80's, but there's still room for
>improvement. Why do so many people install the GNU tools? Could it be
>that they want a more consistent environment across platforms? Could it
>be that there are weaknesses and gaps in the commercial offerings?
no, it's because
1. I dont have a tool to do 'A'
2. the gnu stuff is free
If I got a free copy of sun workshop, I wouldn't need gcc any more.
I suggest you dual-boot solaris x86, and get a wider perspective on life.
Alan Coopersmith
>Read the Makefiles for widely used programs, and you will
>usually see ugly work-arounds for stupid differences between proprietary
>Unices, but no distinction will be made between Linux distributions.
Amazing. I must have imagined all the problems with incompatibilities
between linux'es with different versions of libc ("old" libc vs. GNU
libc, and all the differetn versions of each).
--
________________________________________________________________________
Alan Coopersmith al...@godzilla.EECS.Berkeley.EDU
Univ. of California at Berkeley http://soar.Berkeley.EDU/~alanc/
aka: alanc@{CSUA,OCF,CS,BMRC,EECS,ucsee.eecs,cory.eecs}.Berkeley.EDU
Alan Coopersmith
>Can somebody talk about the differences between sun solaris, BSD, HP/UX,
>DEC/alpha and linux etc. I mean I know that they belong to different
>companies and all. But where is the actual difference? What can solaris do
>that HP/UX can't? Similarly what is the difference between Linux and
>DEC/Alpha or BSD etc? Is there a web page that has a summary of the
>similarities and differences between these types?
First, DEC/Alpha is a CPU type, not an OS. You can run Linux, Compaq
Tru64 (aka Digital UNIX, aka DEC OSF/1), or Windows NT on a Alpha
system.
As for the others, HP/UX will run on a system with the HP PA-RISC CPU,
but not on an Alpha, x86, or Sparc. Compaq Tru64 will run on a DEC
Alpha, but not a PA-RISC, Sparc or x86. Sun Solaris will run on a
Sun Sparc or x86, but not Alpha or PA-RISC. And of course, the free
OS'es (the BSD's & Linuxes) will run on any of the above (well, PA-RISC
support is somewhat limited, but the rest are all there).
HP, Compaq/DEC, and Sun don't see themselves as UNIX OS companies - they
are computer hardware manufacturers who make a version of UNIX for their
boxes so that their customers will buy them (a computer that can't run
any OS is of limited value). Traditionally, they have preferred having
complete control over their OS'es, but that is giving way as companies
like SGI, HP, & IBM embrace linux and open source more.
hac
>
> On Tue, 26 Oct 1999 14:03:13 GMT, hchris...@home.com wrote:
> >...
> >OK, I went too far with the make example, as you and several others
> >point out. But I stand by my point that pressure to differentiate
> >commercial offerings has not always served the UNIX user world well.
> >It's better now than during the 80's, but there's still room for
> >improvement. Why do so many people install the GNU tools? Could it be
> >that they want a more consistent environment across platforms? Could it
> >be that there are weaknesses and gaps in the commercial offerings?
>
> no, it's because
>
> 1. I dont have a tool to do 'A'
> 2. the gnu stuff is free
>
> If I got a free copy of sun workshop, I wouldn't need gcc any more.
>
> I suggest you dual-boot solaris x86, and get a wider perspective on life.
> :-)
>
home. I like Solaris, and I think that it is one of the best UNIX
implementations available. My first UNIX experience was with Version 7
on a PDP-11, then an early BSD on VAX, whatever Sequent calls theirs,
SunOS on 68k, Ultrix, HP/UX. Just to cover the other branch off of the
MULTICS tree, I also used Apollo workstations, which had some
interesting features. They all have flaws, some are better than others,
but I'll take any of them over the non-UNIX type systems I've used.
The UNIX wars of the 80's left a bad taste in my mouth. Users like
myself, and I suspect developers as well, had to live with a lot of crap
as the vendors fought with each other, and tried to lock you into their
solution. It's way better now. I'm having fun with Linux on my home
PC, I get work done on Solaris, and I may install FreeBSD on my old
Pentium motherboard. The users, the customers, have the upper hand now,
and the good bits of all of the variants tend to be widely available
now. These are good times for UNIX. But if I'm a bit wary about
vendors, it's because I remember being burned in the past.
Jon Rouse
>As for the others, HP/UX will run on a system with the HP PA-RISC CPU,
>but not on an Alpha, x86, or Sparc. Compaq Tru64 will run on a DEC
>Alpha, but not a PA-RISC, Sparc or x86. Sun Solaris will run on a
>Sun Sparc or x86, but not Alpha or PA-RISC. And of course, the free
>OS'es (the BSD's & Linuxes) will run on any of the above (well, PA-RISC
>support is somewhat limited, but the rest are all there).
So can I run SunNetManager or HP OpenView on my RedHat linux box?
--
The views expressed are my own, and may not necessarily reflect those of my
employer.
Joerg Schilling
Casper H.S. Dik - Network Security Engineer <Caspe...@Holland.Sun.Com> wrote:
>[[ PLEASE DON'T SEND ME EMAIL COPIES OF POSTINGS ]]
>
>hac <hchris...@home.com> writes:
>
>>usually see ugly work-arounds for stupid differences between proprietary
>>Unices, but no distinction will be made between Linux distributions.
>
>
>much to work on Linux and watch in amazement how programs ported to
>Linux suddenly no longer compiler on the gazillion operating systems they
>used to run on.
If any commercial UNIX vendor would change interfaces in a way Linux does,
it would go bankrupt very soon.
I don't know why interfaces on Linux are changed so often. Maybe the
responsible developers didn't realize that it's an interface that they
are changing....
>It's wrong to assume that all commercial vendors started with the
>same Unix and pissed on it; and when it comes to Unix portability,
>Linux does need some major include file reorganization. (Perhaps it
>could try and follow the UNIX98 standard, even if it doesn't want
>branding)
The Linux include file deaster seems to the main reason why software
that has been developped on/for Linux is hard to port to other UNIX
brands.
Application developers should at least read the UNIX-98 docs before
they work on their programs. This would help to give them a feeling
about what is a common accepted standard and what is a non portable
Linux way of life.
--
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
>>
>> Hi,
>>
>> I am confused by the plethora of unix types out there. I mean what is the
>> somebody talk about the differences between sun solaris, BSD, HP/UX,
>> DEC/alpha and linux etc. I mean I know that they belong to different
>> companies and all. But where is the actual difference? What can solaris do
>> that HP/UX can't? Similarly what is the difference between Linux and
>> DEC/Alpha or BSD etc? Is there a web page that has a summary of the
>> similarities and differences between these types?
>>
>Free or open source software, in contrast, is largely developed by the
>people who want to use the software, not people trying to sell the
>software. Not all users are developers, but the developers are most
>often end users. The users have no reason to compete with each other,
>and every reason to cooperate. So while proprietary UNIX flavors are
>subject to forces that cause them to diverge, the open source
>equivalents are subject to forces that cause them to converge. There
>may be many Linux distributions and BSD flavors, but they have more in
>common than the proprietary UNIX variants. Consensus is a powerful
>force. Read the Makefiles for widely used programs, and you will
>usually see ugly work-arounds for stupid differences between proprietary
>Unices, but no distinction will be made between Linux distributions.
You are completely wrong!
If this was true, why does Linux not follow the UNIX-98 guidlines?
Solaris does!
I would not call Linux open source in a way that I like because
you have no real benefit of having access to the source:
Trying to work on the Linux kernel is fighting against windmills.
I tried to contribute some code enhancements and had no chance because
the Linux kernel team works like the kremlin and did choose a different
solution which was much worse than mine but the contributor of the other
code seems to be liked more then I have been.
If I publish an enhancement or solution for Solaris, I know that I am lost
alone and this is good news ;-)
I would realy likee to see the Solaris source made open to everyone.
It then Sun would pay for some people who look for open source code that
runs on Solaris or extends Solaris (like SCO already does) Linux would
cause less interest than it currently does.
Przemek Bak
> Application developers should at least read the UNIX-98 docs before
> they work on their programs. This would help to give them a feeling
> about what is a common accepted standard and what is a non portable
> Linux way of life.
Where I can find UNIX-98 docs ?
przemol
Joerg Schilling
Philip Brown <phi...@bolthole.no-bots.com> wrote:
>On Tue, 26 Oct 1999 14:03:13 GMT, hchris...@home.com wrote:
>>...
>>OK, I went too far with the make example, as you and several others
>>point out. But I stand by my point that pressure to differentiate
>>commercial offerings has not always served the UNIX user world well.
>>It's better now than during the 80's, but there's still room for
>>improvement. Why do so many people install the GNU tools? Could it be
>>that they want a more consistent environment across platforms? Could it
>>be that there are weaknesses and gaps in the commercial offerings?
>
>no, it's because
>
>1. I dont have a tool to do 'A'
>2. the gnu stuff is free
>
>If I got a free copy of sun workshop, I wouldn't need gcc any more.
And you would get a syntax error message for this piece of code:
switch (i) {
case 1: foo();
case 2:
}
which does not even gives a warning with GCC ;-)
Drazen Kacar
> Where I can find UNIX-98 docs ?
http://www.opengroup.org, I believe. HTML version was free last time I looked.
--
.-. .-. I'm paid for my brain, not my ability not to use it.
(_ \ / _)
| da...@srce.hr
| da...@fly.srk.fer.hr
Gunnar Evermann
> Philip Brown <phi...@bolthole.no-bots.com> wrote:
> >
> >If I got a free copy of sun workshop, I wouldn't need gcc any more.
>
> And you would get a syntax error message for this piece of code:
>
> switch (i) {
>
> case 1: foo();
>
> case 2:
> }
>
> which does not even gives a warning with GCC ;-)
try 'gcc -pedantic-errors' :-) but I completely agree with Philip's
statement above
[F'up set]
Michael Wojcik
[Followups trimmed to comp.unix.solaris, though a redirect to
comp.lang.c is arguably more appropriate.]
In article <7v6s6f$11c$1...@news.cs.tu-berlin.de>, j...@cs.tu-berlin.de (Joerg Schilling) writes:
> In article <slrn81bk16....@shell3.ba.best.com>,
> Philip Brown <phi...@bolthole.no-bots.com> wrote:
> >1. I dont have a tool to do 'A'
> >2. the gnu stuff is free
> >
> >If I got a free copy of sun workshop, I wouldn't need gcc any more.
> And you would get a syntax error message for this piece of code:
>
> switch (i) {
>
> case 1: foo();
>
> case 2:
> }
>
> which does not even gives a warning with GCC ;-)
That's because the authors of gcc don't think people should follow
the ANSI C standard. (Yes, I'm aware you can tell gcc to do so
anyway.)
Read section 6.6. The syntax for a case statement (6.6.1, labeled
statements) is:
case <constant-expression> : <statement>
Your choices for <statement> (from the beginning of 6.6) are:
<statement>:
<labeled-statement>
<compound-statement>
<expression-statement>
<selection-statement>
<iteration-statement>
<jump-statement>
None of those can be empty. The closest you come to allowing
nothing between "case <constant-expression>:" and the end of
the select block are:
case <constant-expression>: (another labeled-statement)
default: (ditto)
; (a minimal expression-statement)
{} (a minimal compound-statement)
(For the first two, of course, you still eventually need
some non-labeled-statement statement before the end of the
select block.)
Now, if the Workshop compiler complained about any of those,
it would be in error. But it doesn't. (Yes, it's happy with an
empty block following a case statement.)
gcc allows all kinds of nonstandard constructions by default. That's
not necessarily a point in its favor.
--
Michael Wojcik michael...@merant.com
AAI Development, MERANT (block capitals are a company mandate)
Department of English, Miami University
Unlikely predition o' the day:
Eventually, every programmer will have to write a Java or distributed
object program.
-- Orfali and Harkey, _Client / Server Programming with Java and CORBA_