SCO market share
Stacey Campbell
>In article <1993Dec03....@sco.com>,
> sta...@socoban.sco.com (Stacey Campbell) writes:
>| Since SCO Unix is installed on about 60 per cent of all Unix-for-Intel
>| platforms...
>
>I'd like to see the proof for this claim. I think it is bogus, unless
>you restrict the term ``Unix-for-intel'' so that it only applies to SCO
>and SCO-look-alikes (whatever that may mean).
>
>But if you can proove that about 60% of all ``Intel platform'' computers
>that are *today* (7-Dec-1993) *running* some sort of Unix-derived OS are
>*running* SCO, I will cheerfully recant.
I of course can't prove anything and may be a victim of my own
company's propaganda, but here's something SCO tacks on to the end
of press releases. It quotes IDC numbers, i.e. an independent
source, and as you can see I got them wrong, we in fact have a bigger
market share than I thought (the terms `accounts for 65 percent of the
market' and licenses installed may or may not be two different things).
Founded in 1979, The Santa Cruz Operation, Inc. (SCO) is the
leading developer and supplier of UNIX systems for Intel
CPU-based computers. According to International Data Corporation,
SCO accounts for 65 percent of the market for UNIX Systems on
Intel platforms, as well as approximately one fourth of all UNIX
Systems installed worldwide. SCO products are available in more
than 80 countries.
If you want to read more of this type of stuff then check biz.sco.announce
(if you have access to it). Obviously this data wasn't compiled today
(7-Dec-1993) but I imagine it's fairly recent.
If it's really important to you, maybe I can find someone in marketing
with the report...
--
Stacey Campbell - sta...@sco.com - uunet!sco!staceyc
Wayne Schlitt
> In article <hastyCH...@netcom.com>, ha...@netcom.com (Amancio Hasty Jr) says:
> | I am interested on what your market Dept. may have to say.
> | Also, I am very interested on any Linux or 386bsd figures that they
> | may have.
>
> I doubt they have anything meaningful. *We* can't even get a reliable count
> of the number of Linux users....
Getting user counts for something like Linux would not be hard to do,
just expensive. "All" that you would have to do is talk to the folks
a Gallup and have them call a couple thousand people and ask them if
the have a computer at work and/or home, and if so, what operating
system they run. Then, take a smaller sample that you can actually go
out and check to see how many people say "DOS" when they are really
running a MAC, etc. Put it all together and extrapolate out. Of
course, doing it for more than just the US would just cost a little
bit more...
Heck, we elect politicans that way, it couldn't be that much worse
for computers.... :->
-wayne
Wayne Schlitt
>
> In <hastyCH...@netcom.com>, ha...@netcom.com (Amancio Hasty Jr) writes:
> >Also, I am very interested on any Linux or 386bsd figures that they
> >may have.
>
> considered as in the Unix-for-Intel *MAARKET* since Linux and 386bsd are
> *FREE*. :-)
You can also *pay* for Linux (and *BSD), if you want to. There are
quite a few companies selling this stuff on floppies, tape and CD-ROM.
One of them, Yggdrasil is claiming that they have shipped ~8,000
Linux CD's within the last 3 months or so. Surveys have shown that
~30-40% of all Linux sites are at work and/or schools. The reliability
of these surveys may be in question, but the indications are that
Linux and *BSD are not just being used in small time outfits and
homes. There are significant businesses doing significant work with
these "free" versions of Unix.
I am curious how significant the market share is. Data collected a
year a go probably won't show much penetration by these systems, but
things are growing quickly, especially the last 3-6 months or so.
> How about Linux and 386bsd, do they come with C Compiler and TCP/IP/NFS?
Yes. Both Linux and *BSD use gcc. *BSD has more stable networking,
but lots of people are using Linux's and the more recent versions are
getting close to *BSD's quality. It should be noted that TCP/IP was
*developed* on BSD, so everyone else must compare thier version to
BSD's. Yellow pages seems to be weak on both Linux and *BSD, or so I
have been told..
-wayne
Mark A. Davis
>For interesting information, read the artical posted in comp.unix.sys5.r4
>about Novell/unixware and creation/use of Emergency Boot Floppys. If Novell
>wants to survive, policies portraied in this artical will surely only
>help SCO ...
And if SCO wants to survive, they are going to have to compete with these
new players by being compatible (IE maintaining ports of apps) and by
being price effective (Unixware is *RAPIDLY* becoming a realistic low-cost
Unix). If Novell includes TCP/IP, Motif, more utilities, and WABI and does
not raise the price (quite a possibility), then SCO is surely on the path
to disaster....
--
/--------------------------------------------------------------------------\
| Mark A. Davis | Lake Taylor Hospital | Norfolk, VA (804)-461-5001x431 |
| Sys.Administrator| Computer Services | ma...@taylor.wyvern.com .uucp |
\--------------------------------------------------------------------------/
Mark A. Davis
>I think it is unrealistic to base a production system on either Linux or 386bsd
>because of the continuing supportability of these products. Most production
>sites do not want to be in the business of being OS developers. Private users
>most certainly will benefit from the availability of these products, and in the
>final anaylysis, all will benefit from advances in functionality.
The reason they are mostly useless in business is because there are no
applications. Applications are what make an OS fly....
Amancio Hasty Jr
I am interested on what your market Dept. may have to say.
Also, I am very interested on any Linux or 386bsd figures that they
may have.
Tnks,
Amancio
--
This message brought to you by the letters X and S and the number 3
Amancio Hasty |
Home: (415) 495-3046 | ftp-site depository of all my work:
e-mail ha...@netcom.com | sunvis.rtpnc.epa.gov:/pub/386bsd/X
Evan Leibovitch
>In article <1993Dec7.1...@rwwa.COM> wi...@rwwa.com writes:
>>| Since SCO Unix is installed on about 60 per cent of all Unix-for-Intel
>>| platforms...
>>I'd like to see the proof for this claim. I think it is bogus, unless
>>you restrict the term ``Unix-for-intel'' so that it only applies to SCO
>>and SCO-look-alikes (whatever that may mean).
>I of course can't prove anything and may be a victim of my own
>company's propaganda, but here's something SCO tacks on to the end
>of press releases. [...]
> According to International Data Corporation,
> SCO accounts for 65 percent of the market for UNIX Systems on
> Intel platforms, as well as approximately one fourth of all UNIX
> Systems installed worldwide.
This sounds right, but remember, it includes Xenix. If you were to
specify the market share of SCO Unix or ODT alone, the numbers would be
considerably smaller.
There are a *lot* of sites out there still running Xenix. For a non-GUI,
non-networked environment, it's a stable environment that's less of a
resource hog than most current Unixoid systems. There's a very good reason
why SCO won't drop Xenix from its product line years after it "upgraded"
to Unix. Xenix still sells.
What's most interesting to me about the 65% number, is not that it's so
high, but that it appears to be falling. I seem to recall it wasn't that
long ago that SCO had about 80% of the market. While that number is nice,
recall that for most of Xenix's early life there was *no* effective
competition. Interactive? Yeah, right.
What I'm more interested in knowing than installed base, is what's
selling *now*. What proportion of the Unix-on-Intel systems shipped in
1993 were SCO, or UnixWare, or Solaris, etc? It doesn't look like *any*
of the vendors are interested in giving out these numbers. If IDC had
any guts, *that's* the stat they should be researching.
--
Evan Leibovitch, Sound Software Ltd., located in beautiful Brampton, Ontario
ev...@telly.on.ca / uunet!utzoo!telly!evan / (905) 452-0504
I can't complain, but sometimes I still do -- Joe Walsh
Mark A. Davis
>In article <1993Dec07....@sco.com> sta...@socoban.sco.com (Stacey Campbell) writes:
>>In article <1993Dec7.1...@rwwa.COM> wi...@rwwa.com writes:
>>>In article <1993Dec03....@sco.com>,
>>> sta...@socoban.sco.com (Stacey Campbell) writes:
>>>| Since SCO Unix is installed on about 60 per cent of all Unix-for-Intel
>>>| platforms...
>>>
>>>I'd like to see the proof for this claim. I think it is bogus, unless
>>>you restrict the term ``Unix-for-intel'' so that it only applies to SCO
>>>and SCO-look-alikes (whatever that may mean).
>>>
>>I of course can't prove anything and may be a victim of my own
>>company's propaganda, but here's something SCO tacks on to the end
>>of press releases. It quotes IDC numbers, i.e. an independent
>>source, and as you can see I got them wrong, we in fact have a bigger
>>market share than I thought (the terms `accounts for 65 percent of the
>>market' and licenses installed may or may not be two different things).
>I am interested on what your market Dept. may have to say.
>Also, I am very interested on any Linux or 386bsd figures that they
>may have.
I think the numbers Stacey/SCO are quoting are for COMMERCIAL Unix systems......
Charlie Li
>In article <1993Dec07....@sco.com> sta...@socoban.sco.com (Stacey Campbell) writes:
>>If it's really important to you, maybe I can find someone in marketing
>>with the report...
>
>Also, I am very interested on any Linux or 386bsd figures that they
>may have.
I'd like to hear the statistics of this too. But I guess these are not
considered as in the Unix-for-Intel *MAARKET* since Linux and 386bsd are
*FREE*. :-)
Another thing: Lately I started to play with Unix-for-Intel (SCO Unix, etc),
surprisingly, C compiler/TCP/IP network components are not included in the
OS system (Most Unix-for-non-Intel come with those components). Those
components are always in separate packages which can add lots of additional
cost. This practice seems to be usual for Intel-based systems (DOS, OS/2).
How about Linux and 386bsd, do they come with C Compiler and TCP/IP/NFS?
What about compiled gcc excuctables for SCO Unix, Linux and 386bsd, any
ftp sites for those? Thanks.
Charlie Li
Larry D Snyder
>I am interested on what your market Dept. may have to say.
you need to remember that SCO includes all licenses -- including Xenix
licenses in these numbers -- and just about *every* unix type person owns
a copy of Xenix (since Xenix was just about the only "unix" option available
years ago).
--
Larry Snyder
la...@gator.oau.org
Stacey Campbell
>ha...@netcom.com (Amancio Hasty Jr) writes:
>>I am interested on what your market Dept. may have to say.
>
>you need to remember that SCO includes all licenses -- including Xenix
>licenses in these numbers
I mentioned this in the original article in comp.windows.x.i386unix, it
turns out that Xenix, while still an important product for us, in fact
makes up a much smaller portion of our sales than I thought.
Here's the three tables from the IDC report, The Unix Systems Market,
1992-1997, David Card, September 1993, IDC #7867, that will be of
interest. It's the data in these tables that marketing used to calculate
that we have 65 per cent of the Unix-for-Intel market, and just under
a quarter of all installed Unix systems (really 22 per cent).
Value* of Worldwide Unix System Shipments by Architecture, 1992
Solaris/Sparc 16.9 per cent
HP-UX/HP-PA 12.2
AIX/Power 10.0
SCO/Intel 6.0
Irix/MIIPS 4.1
Intergraph/Clipper 3.1
NCR Intel 2.8
Ultrix/MIPS 2.7
Unicos/Cray 2.7
Others 39.5
I guess that Solaris/Intel is in `others' and its share is less than 2.7
per cent. * - I haven't the foggiest what `Value' means in this context.
Worldwide Unix System Shipments by Product, 1992
Solaris 1 17.5
Solaris 2 6.2
SCO Unix and ODT 14.3
SCO Xenix 4.0
HP-UX 10.0
AIX 6.8
Interactive 3.0
Ultrix 3.0
NeXTstep 2.7
Irix 2.7
AT&T SVR4 (Intel) 2.2
Others 27.6
Market Share By Installed UNITS of UNIX Systems
SCO 21.18
SunOS 18.65
HP-UX 8.43
Interactive 4.24
Ultrix 2.5
AIX 4.49
Other Intel 2.94
Other Sys 16.21
Other pro 11.98
Other tot[oa] 8.36
The last three `other' names are truncated in the report, rather than
guess what they are, I just copied out as much as is legible.
The IDC article has a bunch more stuff, lots of analysis, etc. An
interesting read.
>-- and just about *every* unix type person owns
>a copy of Xenix (since Xenix was just about the only "unix" option available
>years ago).
Larry! Since you're one of the oldest (no offense intended!) Unix type
people I know, could this mean you are a closet owner of a SCO product?!
Amancio Hasty Jr
>In <hastyCH...@netcom.com>, ha...@netcom.com (Amancio Hasty Jr) writes:
>>In article <1993Dec07....@sco.com> sta...@socoban.sco.com (Stacey Campbell) writes:
>>>If it's really important to you, maybe I can find someone in marketing
>>>with the report...
>
>considered as in the Unix-for-Intel *MAARKET* since Linux and 386bsd are
>*FREE*. :-)
>
>Another thing: Lately I started to play with Unix-for-Intel (SCO Unix, etc),
>surprisingly, C compiler/TCP/IP network components are not included in the
>OS system (Most Unix-for-non-Intel come with those components). Those
>components are always in separate packages which can add lots of additional
>cost. This practice seems to be usual for Intel-based systems (DOS, OS/2).
>How about Linux and 386bsd, do they come with C Compiler and TCP/IP/NFS?
The FreeBSD or NetBSD distribution includes binaries and sources for
gcc/tcp/ip/NFS/X cdrom support. NetBSD offers the ability to run
BSDI binaries and has dynamic shared libraries also it is being
ported to different platforms.
Oops I forgot not mentioned no motif.
freebsd at freebsd.cdrom.com:/pub/FreeBSD
netbsd at agate.berkeley.edu:/pub/NetBSD
Currently telnet to this machine from my home machine running netbsd.
My home connection to the internet is via slip. At home my entire
X binary distribution was compiled with shared libraries.
Enjoy,
Tim Ruckle
In article <CHpyE...@telly.on.ca> ev...@telly.on.ca (Evan Leibovitch) world:
} [...]
} selling *now*. What proportion of the Unix-on-Intel systems shipped in
} 1993 were SCO, or UnixWare, or Solaris, etc? It doesn't look like *any*
} of the vendors are interested in giving out these numbers. If IDC had
} any guts, *that's* the stat they should be researching.
Evan, I doubt that IDC lacks the "guts" to report these numbers--their whole
business is based on providing this sort of impartial and objective data.
1993 is not quite over as yet, and IDC will need a little time before they
can compile and publish their numbers for this year. But I'm sure they are
researching those stats.
But if the trade rag reports of UnixWare shipments are to be believed, SCO
may very well be quite interested in giving out those numbers. ;^)
We shall see. To my mind a more interesting breakdown will be the server
to desktop ratio of the licenses--otherwise we compare apples to oranges,
no?
Regards,
Tim
--
The ship was cheered, the harbor cleared,
Merrily did we drop
Below the kirk, below the hill,
Below the lighthouse top.
--Coleridge
Tim Ruckle
In article <1993Dec08....@sco.COM> dav...@sco.COM (Dave Gurr) writes:
}
} In 1992, SCO shipped around 18% of all UNIX licenses, slightly
} behind the leading shipper (Sun) on 24% (source: IDC 1993)
}
} Some other interesting stats:
}
} In 1992 SCO was responsible for around 22% of the installed
} base of all UNIX systems worldwide - more than any other vendor,
} including Sun on around 19% (source: IDC 1993)
}
} An impartial observer may wish to comment on the likely number of connected
} users for the installed/shipped systems above, and thus the likely user
} populations existing, and added to.
Hi Dave,
I heard that SCO Systems had an average of 7.7 users per system a few years
back (as reported by Dataquest). With the rise of SCO MPX and SCO Open
Server Systems (and the vast increase in performance of industry-standard
hardware) this may very well be higher by now.
In any case, though I can't claim to be a totally impartial observer, it
stands to reason that if Sun Systems have a greater workstation to server
ratio than SCO does, then their total number of user "seats" will be less
(and visa versa).
Is this what you were getting at?
} SCO only sells system software ... but if you take into account
} all the system hardware and peripherals that our systems are sold
} with, in 1992 we were responsible for around $2bn worth of downstream
} sales (more than $3bn if you include the application software sales)
} (Source: Aberdeen Group 1993)
Hmm, this is probably only interesting in as much as somebody tries to use
bogus statistics to try and paint SCO as a minor player in the Open Systems
market: if you go ahead and plot total sales dollars then SCO is hardly a
blip on the screen compared to Sun, HP, and IBM.
But SCO is a vendor-independent software company, so those billions in system
hardware sales "downstream" don't count.
There are lies, dang lies, and statistics. I'm not sure what this thread has
to do with the price of XENIX in China (where SCO found itself with a huge
market share, BTW, before they even sold a single copy ;^). And while I do
trust those IDC numbers, I don't see how the relative market share stats are
really at issue here.
If we can't all grow the total market then we're all in trouble. And the
worst thing in the world would be to see one of our main "competitors" go
under and give up all their market share to one of us.
(the reason for this is left as an exercise for the reader--please show all
work... |')
Regards,
Tim
--
unless statistics lie he was
more brave than me:more blond than you.
--e.e. cummings
Charles M Richmond
>Bruce_M...@qad.com writes:
>>I think it is unrealistic to base a production system on either Linux or 386bsd
>>because of the continuing supportability of these products. Most production
>>sites do not want to be in the business of being OS developers. Private users
>>most certainly will benefit from the availability of these products, and in the
>>final anaylysis, all will benefit from advances in functionality.
>The reason they are mostly useless in business is because there are no
>applications. Applications are what make an OS fly....
...and the reason that there are no applications , is that the gcc "copyleft"
leaves no commercial incentive to develop applications. I guess that the
appearance of a commercial compiler , or at least headers and libraries
would go a long way toward starting the ball rolling.
Charlie
--
*****************************************************************************
* Charles Richmond Integrated International Systems Corporation *
* c...@world.std.com *
* Specializing in UNIX, X, Image Processing, and Communications. *
Brian Connell
First of all, Hi Larry. It's been a while since you've posted on here. I thought you
swore never to post here again :-).
Also, what's wrong with counting the Xenix licenses? Without Xenix, we probably
wouldn't have such a strong desktop UNIX market as we have today. The reason that
Xenix was the only un*x option available is because all the big players had written
off the possibility of a market on the desktop. Look at them now!! What a mess!
Everybody is now madly scrabbling away for a market share, including all those SVR4
versions, while SCO have kept cool, pedalled slowly, and ensured that the most commonly
used words that people describe their product with are words like "solid", "stable",
"reliable" and "large installed base".
BTW. It would also be very interesting to see the volume of licenses sold in 1993 by
all the desktop unix vendors. I'd say that SCO is at No. 1, with a long gap to No. 2.
Regards,
Brian.
* . + . *
-=-=-=-=-=-=-=-=-=-=-=-=-=-=--=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= + `. ' * ' + =-
Brian Connell - (bcon...@ingres.com)| ~~~~~ '* ' * ' . + . . ` `
Tel: +353-1-662-0133 | =|###| . '. ' ' o ^ o ` `+
Treasury Building, || |###| G U I N N E S S + ( |\O/| ) * +
Lower Grand Canal Street, | =\###/ . ' ' +++( \|/ )+ +
Dublin 2, IRELAND. | --- . * + + ( /|\ )++ +
-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= ++++( ||| )- + -
+ ~ ~
Rick Kelly
>+---------------
>| I am interested on what your market Dept. may have to say.
>| Also, I am very interested on any Linux or 386bsd figures that they
>| may have.
>
>I doubt they have anything meaningful. *We* can't even get a reliable count
>of the number of Linux users....
Just look at the Linux serial number data base..........
--
Rick Kelly r...@rmkhome.com r...@bedford.progress.com
Harry Skelton
>And if SCO wants to survive, they are going to have to compete with these
>new players by being compatible (IE maintaining ports of apps) and by
>being price effective (Unixware is *RAPIDLY* becoming a realistic low-cost
>Unix). If Novell includes TCP/IP, Motif, more utilities, and WABI and does
>not raise the price (quite a possibility), then SCO is surely on the path
>to disaster....
>
SCO has always held a mirror to itself and always failed to view what other's
in the Unix industry are doing. From trying to get SCO to use X, to trying
to get SVR4 conformance (which I hear is still very lacking), to getting a
working NFS suite, etc, etc etc...
A while back, before the current Marketing VP, I was told by the 'then' VP
that SCO has no interest in following any other company. That they had a
market that was unique and the users were perfectly happy with what SCO was
offering.
This is true for someone that knows little more than how to turn on the
computer and start the application a vendor has loaded on there for them.
But in the case of the technically minded, SCO's managment, sorta like
Novell, has a DOS'ism policy that requires Unix to conform more to DOS than
DOS to Unix. But Novell has high marks for getting away from that image,
although at a snail's pace.
BTW: Could someone send me the release notes to the current release of SCO
Unix? That and the glossies? Thanks.
--
Harry Skelton - 1848 Beaver Dam Lane - Marietta, Georgia - 30062
404-590-7100 or 800-366-8181 Work -- 404-578-8085 Home
ske...@jdp.dragon.com
Mark A. Davis
>ma...@taylor.wyvern.com writes:
>>
>> >I think it is unrealistic to base a production system on either Linux or 386bsd
>[text deleted]
>> The reason they are mostly useless in business is because there are no
>> applications. Applications are what make an OS fly....
>Not this debate again, I am new to this news group so I am not sure how
>many times this issue of free Operating Systems has been discussed,
[lots deleted]
You did not talk about applications, though. That was my only point in
the above posting. Linux cannot run any of the commerical applications,
making it a non-choice for most business use. This could change, however....
Eric S. Hvozda
>
> Founded in 1979, The Santa Cruz Operation, Inc. (SCO) is the
> leading developer and supplier of UNIX systems for Intel
> CPU-based computers. According to International Data Corporation,
> SCO accounts for 65 percent of the market for UNIX Systems on
> Intel platforms, as well as approximately one fourth of all UNIX
> Systems installed worldwide. SCO products are available in more
> than 80 countries.
Proud to say, I'm not part of this statistic; NetBSD is is free and that's
for me!
--
Ack!
Creek, not creek; Pop not soda; Car needs washed...
Ray Jones
>wa...@backbone.uucp (Wayne Schlitt) writes:
>>One of them, Yggdrasil is claiming that they have shipped ~8,000
>>Linux CD's within the last 3 months or so. Surveys have shown that
>>~30-40% of all Linux sites are at work and/or schools. The reliability
>>of these surveys may be in question, but the indications are that
>>Linux and *BSD are not just being used in small time outfits and
>>homes. There are significant businesses doing significant work with
>>these "free" versions of Unix.
>these companies who produce the number surveys are paid for their work
>by the unix vendors. I would rather doubt if they would favor (or report)
>the free variations of Unix available since they don't pay for these reports
>or report contents
I have to disagree with Larry's statement. I have spent a significant amount
of time doing those surveys and writing that type of report (for McGraw-Hill
and others). I have also been one of those marketeers that paid for those
reports.
If a company is makeing a significant investment in the design, development
and marketing of a product, it is vital to know who is doing what in that
market. The more acurate, and trueful the information the better.
If you are threatened by a "free" product (Linux), that is critical
information.
BTW, in this area (I can only speak of the groups that I have actually seen)
there is a great interest in Linux. However, all but two of the ones I have
seen are being used by individuals. Most of the companies we deal with are
interested in looking at Linux, for the personal use of the individuals (at
home) but not for the mission critical company use.
--
INTERNET: r...@Celestial.COM | The probability of one or more
Ray A. Jones; Celestial Software | spelling errors in this missive
8545 S.E. 68th Street | approaches unity. If this bothers you,
Mercer Island, WA 98040;(206) 236-1676 | run it through your spell checker!
Amancio Hasty Jr
>In <CHrq7...@gator.oau.org> la...@gator.oau.org (Larry D Snyder) writes:
>
>>wa...@backbone.uucp (Wayne Schlitt) writes:
>
>there is a great interest in Linux. However, all but two of the ones I have
>seen are being used by individuals. Most of the companies we deal with are
>interested in looking at Linux, for the personal use of the individuals (at
>home) but not for the mission critical company use.
Well, I work in Silicon Valley, and most invididuals that I have spoken to
which want Unix at home are either running *bsd or Linux. Granted is not
an official survey by any means of the imagination.
Bela Lubkin
>be...@sco.com wrote:
>> I do not believe these prices are realistic. In the US, one does not
>> but ODT 3.0 Personal System + a "server upgrade"; you simply buy SCO
>> Open Server Enterprise System (16-user). The price ratios between the
>> US product line are similar to what you show above, which makes me think
>> that you are actually installing OSES-16 on top of ODT PS. You *can* do
>> that, but you're just wasting money. So eliminate the first item from
>> the above, leaving the totals at A$17640, $11677.68.
>
> - The prices I quoted were taken directly from my price list
> provided to me by the SCO distributor 4 days ago, believe me
> these are accurate. I will divulge the name of the company over
> the net as I feel that this is not good form.
>
> - The original recommendation made by the SCO distributors was to
> buy the Personal system and upgrade to the Server, this has been
> consistant practice since ODT 1.0. I have put forward your
> statement to them and they have not got back to me yet.
First, "thank you" for posting my private email to the net. This is
certainly a fine example of good form.
Second, I am not your SCO distributor and I feel little personal
responsibility for their alleged poor business practices. In fact, in
order to get a multiuser ODT 1.0-based system you did have to buy a
"Personal System" and a "Server Upgrade". That is not true of the 3.0
product.
> - Even if you save AU$2725.00 / US$1803.95 is it worth it
> considering the fact that the machine I compared to it was far
> more superior in performace, support options and applications/
> development tools available.
Third, you are deliberately ignoring one of the real advantages of running
an OS on Intel-based hardware, which is that you don't have to buy the
hardware from the software vendor. You can't very well run HP/ux for HP
Precision Architecture on a third-party machine because (as far as I know)
there are no "HP PA clones". Yet you choose to buy your Intel hardware
from the same source that will sell you the software. Chances are that
hardware will be of fairly good quality, but at a substancial premium in
price. If you choose to investigate the market yourself and buy
appropriate hardware, you can lower the price of the ODT system
considerably. Note that I am *not necessarily recommending this*, simply
pointing it out as an avenue by which the ODT system provides a softer
price than a single-vendor solution.
>> Most net.software downloads, builds and installs without much trouble on
>> ODT 3.0 + ODT DS 3.0.
>
>> I dispute the statement "difficult to port software to". ODT 3.0 is
>> much better about this than earlier releases; are you working from
>> current information?
>
>Granted, I am referring to ODT 2.0 and I am still deciding whether to update
>to 3.0 or write off my investment.
>
>> I dispute "complexity of UNIX env", only in that you did not tag HP/ux
>> with the same "disadvantage".
>
>But SAM doesn't trash your SCSI hard disk or not set an option in
>some critical configuration file.
I assume SAM is an HP system administration tool.
What ODT software "trashed" your SCSI disk? Did you provide SCO with a
report of the problem? What option was left unset in a "critical
configuration file"? This is meaningless innuendo as far as I'm
concerned.
>> Finally, I completely dispute "released only 2.5 years ago". ODT 1.0
>> was released in mid to late 1989, 4 years ago. SCO UNIX 3.2.0 was
>> released in early 1989.
>
>Thats funny, when SCO sent the product out to our development project, we
>were the first people to receive the "Production Release" of the product and
>that was Mid 1990. Regardless of how long, the issue is that there are more
>mature and reliable UN*Xes out there.
There may have been a delay in introducing the product in Australia. I do
not know. Note that SCO has been in the UNIX business (with XENIX) for
over 10 years. SCO UNIX is not SCO XENIX but it derives a lot of maturity
from its ties to XENIX. (Uh-oh, I can hear the flames already...)
>> SCO UNIX has scalability through MPX, which is scalable hardware. A
>> 4-processor Pentium box competes well for CPU power with a high-end
>> single processor HP PA box.
>
>This is where I get a little annoyed, this statement is based on what
>commercial implementation?
>
>This is the same statement made by the marketing people here in Australia.
>When I ask them to take me to the site and lets discuss implementation and
>perfomance issues, they reply , "OH! we need to wait for the release of
>the Pentium", that is absolutely no use to our clients, they need proven
>working solutions. When the Pentium is available in Australia WITH PCI
>bus we may have something to discuss on this issue, until then tell me
>what I can deliver NOW!
I am not 100% up to date on what's on the market. I believe that
Acer/Altos, AST, Compaq, Tricord, and several other manufacturers have
multiprocessor Pentium machines on the market; some of them support 4 or
more processors. But perhaps I should have said that a 4-processor 486/66
box is on par with a high-end single processor HP PA. There are many
486/66 MP machines on the market.
PCI is not a requirement! It's an up-and-coming technology; certainly it
would be *nice* to have in your new server but that does not make it
necessary.
>> (BTW, it's "Linux", based on the name of the author, Linus Torvaalds. I
>> imagine he does not enjoy seeing the name of his creation distorted on
>> the net.)
>
>Good to see you were able to add something substantial to the discussion!
>
>Come back to me with FACT.
Wooo, hit me harder! Your grasp of "FACT" doesn't seem any stronger than
mine; what gives you the right to be so smug?
Let's take this useless bickering back to email. I prefer to spend net
bandwidth on solving substantative technical problems and useful
speculation.
>Bela<
Fred Rump from home
>The difference is that almost all of the Solaris and Unixware sales are to
>corporate sites that are reviewing the concept of Unix on Intel, or for pilot
>projects. Almost none of these sales are for production implementation.
On the other hand, SCO is moving large quantity sales into that same corporate
environment bypassing its somewhat its base of small end-users. I'm told that
1000 quantity sales are not that much out of the ordinary any more. Revenue
from MPX add-on sales may also be adding up.
Fred
--
"Dreifach ist der Schritt der Zeit: Fred Rump f...@icdi10.compu.com
Zo"gern kommt die Zukunft hergezogen, 26 Warren St, Beverly, NJ 08010
Pfeilschnell ist das Jetzt entflogen, 609-386-6846
Ewig still steht die Vergangenheit." (Schiller)
Fred Rump from home
>What's most interesting to me about the 65% number, is not that it's so
>high, but that it appears to be falling. I seem to recall it wasn't that
>long ago that SCO had about 80% of the market. While that number is nice,
>recall that for most of Xenix's early life there was *no* effective
>competition. Interactive? Yeah, right.
Obviously new entrants into the market will reduce marketshare for the early
leaders.
>What I'm more interested in knowing than installed base, is what's
>selling *now*. What proportion of the Unix-on-Intel systems shipped in
>1993 were SCO, or UnixWare, or Solaris, etc? It doesn't look like *any*
>of the vendors are interested in giving out these numbers. If IDC had
>any guts, *that's* the stat they should be researching.
What SCO is presenting is a $200,000,000 annual sales revenue. Devide the
numbers and a fairly good OS sales statistical scenario should result.
Jim Sullivan
>My point however regardless of SCO market share is, most of our clients see
>SCO as not even in the game because of price/performance, and long term
>who cares if your market share is even 100% of the Intel market if the Intel
>architecture is increasingly been seen as not been able to deliver the
>goods compared to the RISC based commercially proven alternatives.
Might I suggest that you look at some of the benchmarks that exist for
the UNIX environment. Yes, I know that benchmarks aren't real world,
but they do allow us to compare systems (and I mean systems, not pieces
of silicon) with a standard yardstick (or meterstick in canada :-).
Last I looked, (October 1993) Intel systems from Compaq, SCO and Oracle
were setting price/performance standards with the TPC-A benchmark.
When I looked at the recent AIM report I got last week, there were
Intel systems (with most of them SCO Intel systems) is almost every
Best catagory, from <$15,000 to >$500,000.
--
Jim Sullivan
SCO Canada, Inc. Toronto, Ontario If you are lucky, you'll experience
j...@sco.com, ...!uunet!scocan!jim only 80 individual seasons. Enjoy
416 922 1937 (8397 for fax) them while you can.
Stuart Lynne
Brandon S. Allbery <b...@kf8nh.wariat.org> wrote:
}In article <hastyCH...@netcom.com>, ha...@netcom.com (Amancio Hasty Jr) says:
}+---------------
}| Well, I work in Silicon Valley, and most invididuals that I have spoken to
}| which want Unix at home are either running *bsd or Linux. Granted is not
}| an official survey by any means of the imagination.
}
}Most home users have better uses for the money it would take to buy SCO,
}Interactive/SunSoft, Esix, etc. Like paying bills....
This is a fact that is *not* lost on application software vendors. People
who don't like to pay money for the OS also don't usually like to spend
money on application software. So the vendors don't worry about the number
of users of Linux/Minix/*bsd/?? simply because they don't view them as
potential customers.
*bsd and Linux make a great hobby. If that's what you want to do at night
it's probably just as interesting as ham radio or model trains :-)
But I think that it has about the same effect on the rest of the industry as
ham radio or model train enthusiats. Well maybe a little more.
If I was doing this as a hobby I would certainly be using *bsd. I think it
would be a *lot* of fun to do some drivers for it. Unfortunately I'm in the
unfortunate position of having to make a living do this, and reserve my
spare time for family and other recreations (like skiing, whistler got a
foot of snow last night!!!). So I concentrate my programming energies where
I have the highest probability of getting people to pay me for what I do.
Which right now is SCO (on intel at least, also get the very occasional
query about UnixWare, yuk!)
--
Stuart Lynne <s...@wimsey.com> ........... 604-936-8649(voice) 604-937-7718(fax)
Installable Software for SCO UNIX ............ ftp.wimsey.com:~ftp/pub/wimseypd
Internet SCO UNIX troubleshooting ......................UNIX Facsimile Software
A Wizard of Earth C
>*bsd and Linux make a great hobby. If that's what you want to do at night
>it's probably just as interesting as ham radio or model trains :-)
>
>But I think that it has about the same effect on the rest of the industry as
>ham radio or model train enthusiats. Well maybe a little more.
There are a number of changes prototyped on *BSD that are going into a
future SVR4 release and/or are available publically already. Besides
rather arcane kernel changes, there are sound card drivers and DOS and
ISO9960 File system support (to name a couple).
*BSD is my hobby because I like to program, and programming what I want
to program as opposed to what I am told to program is my way of not
burning out.
That doesn't mean that nothing I do can be rolled over into my job,
either as code or simply as experience and knowledge gained.
You'd be suprised at the number of professional kernel hackers that
have Linux or *BSD installed on some machine.
Terry Lambert
te...@cs.weber.edu
---
Any opinions in this posting are my own and not those of my present
or previous employers.
Lee Penn
>> Might I suggest that you look at some of the benchmarks that exist for
>> the UNIX environment. Yes, I know that benchmarks aren't real world,
>> but they do allow us to compare systems (and I mean systems, not pieces
>> of silicon) with a standard yardstick (or meterstick in canada :-).
>> Last I looked, (October 1993) Intel systems from Compaq, SCO and Oracle
>> were setting price/performance standards with the TPC-A benchmark.
>> When I looked at the recent AIM report I got last week, there were
>> Intel systems (with most of them SCO Intel systems) is almost every
>> Best catagory, from <$15,000 to >$500,000.
>We deal with medium to large comanies and Governmet bodies all of which
>do not by from specifiactions, all of which by from reviewing reference
>sites.
>(ri...@uninet.com.au) Melbourne Victoria Australia 3001
> Phone: Intl.+(613) 305 5649
I agree more with Jim. Every large corporation type of bid has a
requirement for TPC benchmarks. The current popular one seems to be
TPC-C which is supposed to be tougher on those who would tune things
specifically for the benchmark.
Benchmarks are important. Without them, they often don't even bother
with reference sites.
================================================================================
Lee Penn, DLP Technologies, Inc. Voice: 513 232-7791
SCO Advanced Product Center FAX: (513) 232-7801
7444 Jager Ct.
Cincinnati, OH 45230-4344 email: l...@dlpco.COM
Steve Sheldon
>In <931210153...@wattle.uninet.com.au> Rick Ruggiero <ri...@uninet.com.au> writes:
>>My point however regardless of SCO market share is, most of our clients see
>>SCO as not even in the game because of price/performance, and long term
>>who cares if your market share is even 100% of the Intel market if the Intel
>>architecture is increasingly been seen as not been able to deliver the
>>goods compared to the RISC based commercially proven alternatives.
>Might I suggest that you look at some of the benchmarks that exist for
>the UNIX environment. Yes, I know that benchmarks aren't real world,
>but they do allow us to compare systems (and I mean systems, not pieces
>of silicon) with a standard yardstick (or meterstick in canada :-).
>Last I looked, (October 1993) Intel systems from Compaq, SCO and Oracle
>were setting price/performance standards with the TPC-A benchmark.
>When I looked at the recent AIM report I got last week, there were
>Intel systems (with most of them SCO Intel systems) is almost every
>Best catagory, from <$15,000 to >$500,000.
A lot depends on what you are doing, and what kind of benchmarks you are
comparing.
From the stuff I have done, and the benchmarks I have tried, I've found
that a 486DX2/66 has FPU performance similar to our DECstation 5000/200.
This machine uses an older MIPS R3000 chip, at like 33 Mhz or so.
But the newer RISC chips, such as the R4400, or the Alpha provide even
greater FPU performance.
A lot depends on what you are doing. For scientific applications, you need
the FPU power of the RISC chips.
Although I have seen some benchmarks recently also seem to show the Alpha
machines are pretty fast at database operations as well.
--
Steve Sheldon [These are my own opinions]
Iowa State University ICSS Resource Facility by day
she...@iastate.edu ProMap, Inc. by night
Amancio Hasty Jr
>percentages. There are tens of thousands (at least) of copies of
>Linux in use out there.
>
Well someone posted that for 1992 SCO sales revenue was $200,000,000
So lets assume that each unix system cost $500 (way below wholesale
price :-) )
This means that SCO shipped approximately 400,000 unix systems for the
year 1992.
Now who wants to bet that the combined linux and *bsd installed base
for the 1993 is greater than 400,000 :->
Rev. Michael P. Deignan
> Yes, Linux comes with gcc, ++, mail, TCP/IP, CD-ROM support,
> soundboard support, XWindows, on and on. And it's free.
Unfortuntely, it doesn't support a fraction of the device drivers
that are available on the market for other devices, such as a
Always Technology SCSI board, or a Specialix SIO card.
MD
--
-- Michael P. Deignan, KD1HZ -
-- Internet: kd...@anomaly.sbs.com - I never tell the truth, because I
-- UUCP: ...!uunet!anomaly!kd1hz - I don't believe that there is such
-- AT&TNet: 401-273-4669 - a thing...
Dave Gurr
>
>
> In article <1993Dec08....@sco.COM> dav...@sco.COM (Dave Gurr) writes:
> }
> } In 1992, SCO shipped around 18% of all UNIX licenses, slightly
> } behind the leading shipper (Sun) on 24% (source: IDC 1993)
> }
> } Some other interesting stats:
> }
> } In 1992 SCO was responsible for around 22% of the installed
> } base of all UNIX systems worldwide - more than any other vendor,
> } including Sun on around 19% (source: IDC 1993)
> }
> } An impartial observer may wish to comment on the likely number of connected
> } users for the installed/shipped systems above, and thus the likely user
> } populations existing, and added to.
>
> Hi Dave,
>
> I heard that SCO Systems had an average of 7.7 users per system a few years
> back (as reported by Dataquest). With the rise of SCO MPX and SCO Open
> Server Systems (and the vast increase in performance of industry-standard
> hardware) this may very well be higher by now.
>
> In any case, though I can't claim to be a totally impartial observer, it
> stands to reason that if Sun Systems have a greater workstation to server
> ratio than SCO does, then their total number of user "seats" will be less
> (and visa versa).
>
> Is this what you were getting at?
Yup. Though I've no independent figures on how Sun compare - just a gut
feeling.
> } SCO only sells system software ... but if you take into account
> } all the system hardware and peripherals that our systems are sold
> } with, in 1992 we were responsible for around $2bn worth of downstream
> } sales (more than $3bn if you include the application software sales)
> } (Source: Aberdeen Group 1993)
>
> Hmm, this is probably only interesting in as much as somebody tries to use
> bogus statistics to try and paint SCO as a minor player in the Open Systems
> market: if you go ahead and plot total sales dollars then SCO is hardly a
> blip on the screen compared to Sun, HP, and IBM.
Absolutely. Unfortunately, it's all too often that assessments are made of
which are the major companies and driving forces in the industry based solely
on company revenues.
> But SCO is a vendor-independent software company, so those billions in system
> hardware sales "downstream" don't count.
True. We don't get the dollars (damn shame if you ask me :-), but these
figures are useful in getting a view of the wider picture.
--
The opinions expressed above are those of the author's cat, probably not
those of the author, and almost certainly NOT those of SCO Ltd. Miaouw.
dav...@sco.COM Dave Gurr, Market Development Manager-Desktops, SCO Ltd
Dan Mattrazzo
>I'd like to hear the statistics of this too. But I guess these are not
>considered as in the Unix-for-Intel *MAARKET* since Linux and 386bsd are
>*FREE*. :-)
>
>Another thing: Lately I started to play with Unix-for-Intel (SCO Unix, etc),
>surprisingly, C compiler/TCP/IP network components are not included in the
>OS system (Most Unix-for-non-Intel come with those components). Those
>components are always in separate packages which can add lots of additional
>cost. This practice seems to be usual for Intel-based systems (DOS, OS/2).
>How about Linux and 386bsd, do they come with C Compiler and TCP/IP/NFS?
>What about compiled gcc excuctables for SCO Unix, Linux and 386bsd, any
>ftp sites for those? Thanks.
>
>Charlie Li
>
Yes, Linux comes with gcc, ++, mail, TCP/IP, CD-ROM support,
soundboard support, XWindows, on and on. And it's free.
-------------------------------------------------------------------------------
Dan Mattrazzo
dcm...@ritvax.isc.rit.edu
Mastering that Parallel thing
Graduate Studies
Computer Science
Rochester Institute of Technology
Chris Gayler
>
>Getting user counts for something like Linux would not be hard to do,
>just expensive. "All" that you would have to do is talk to the folks
>a Gallup and have them call a couple thousand people and ask them if
>the have a computer at work and/or home, and if so, what operating
>system they run. Then, take a smaller sample that you can actually go
>out and check to see how many people say "DOS" when they are really
>running a MAC, etc. Put it all together and extrapolate out. Of
>course, doing it for more than just the US would just cost a little
>bit more...
This wouldn't work unfortunately because the population of Linux users
is too small in the general population. Assuming a general population of
250 million, your 2K sample size would show the Linux population shifting
by 125,000 for every positive response, a high margin of error. If just
5% of that 2K sample lied or misunderstood the question (a Lexus? Sure
I own a Lexus ...), your report would show over 12.5 Million Linux users
where there might really be only 100,000.
You'd have to increase the sample size beyond belief, making it expensive
beyond belief.
- Chris
Dave Gurr
[ stuff deleted ]
> What I'm more interested in knowing than installed base, is what's
> selling *now*. What proportion of the Unix-on-Intel systems shipped in
> 1993 were SCO, or UnixWare, or Solaris, etc? It doesn't look like *any*
> of the vendors are interested in giving out these numbers. If IDC had
> any guts, *that's* the stat they should be researching.
They have ...
In 1992, SCO shipped around 18% of all UNIX licenses, slightly
behind the leading shipper (Sun) on 24% (source: IDC 1993)
Some other interesting stats:
In 1992 SCO was responsible for around 22% of the installed
base of all UNIX systems worldwide - more than any other vendor,
including Sun on around 19% (source: IDC 1993)
An impartial observer may wish to comment on the likely number of connected
users for the installed/shipped systems above, and thus the likely user
populations existing, and added to.
The IDC quote of 65% of all Intel UNIX license shipments worldwide
has been quoted already, but equally interesting is the quote
that SCO shipped *76%* of all Intel UNIX licenses in Europe in 1992
(source: Dataquest 1993)
SCO only sells system software ... but if you take into account
all the system hardware and peripherals that our systems are sold
with, in 1992 we were responsible for around $2bn worth of downstream
sales (more than $3bn if you include the application software sales)
(Source: Aberdeen Group 1993)
--
A Wizard of Earth C
>dcm...@ritvax.isc.rit.edu (Dan Mattrazzo) writes:
>
>> Yes, Linux comes with gcc, ++, mail, TCP/IP, CD-ROM support,
>> soundboard support, XWindows, on and on. And it's free.
>
>Unfortuntely, it doesn't support a fraction of the device drivers
>that are available on the market for other devices, such as a
>Always Technology SCSI board, or a Specialix SIO card.
Are you referring to DOS device drivers (which don't belong in these
news groups) or are you referring to other OS's having more device
support?
I can assure you that most commercial UNIX implementations THAT HAVE
BEEN RELEASED don't:
o Support QIC-40/QIC-80 drivers unless they have rewritten
clock.c for sufficient timer resoloution (commercial UNIX
timer resoloution is 10ms, and this is INSUFFICIENT).
o Don't support mounting of ISO9960 CD ROMs unless they have
installed the PD "File System Survival Kit".
o Don't support mounting of DOS partitions unless they have
installed the PD "File System Survival Kit"... AND rewritten
their boot code AND dealt with the drive translation issues
AND added support for DOS extended partitions.
o Support sound cards (NeXTStep isn't UNIX, sorry).
o Support Diamond video cards that have proprietary clock
programming interfaces (unless they buy an add-on server
like Pittsburg Powercomputing or MetroLink).
o Support Adaptec 2742 (or AIC-7770 embedded) SCSI controllers.
o Support Mitsumi CD ROM drives.
o Support CD ROM drives on "SoundBlaster" interfaces.
o Directly support EISA devices (example: SCSI DMA not using
bounce buffers when more than 16M of memory exists).
o Support WABI.
o Support Intel video capture boards.
o Cost the same as (or less than) DOS+Windows.
What UNIX systems *do* support the Always Technology SCSI board?
What UNIX systems *do* support the Specialix SIO card? (If it is a
board without an on-board processer, I guarantee you one of Linux
or *BSD supports it).
A Wizard of Earth C
>The FreeBSD or NetBSD distribution includes binaries and sources for
>gcc/tcp/ip/NFS/X cdrom support. NetBSD offers the ability to run
> BSDI binaries and has dynamic shared libraries also it is being
> ported to different platforms.
>
>Oops I forgot not mentioned no motif.
Actually, several Motif and Motif-like and Motif-clone implementations are
available for one or both systems (4 actually, although one is a version
1.1 and one is a version 1.1.2 from the same source). Information follows:
Terry Lambert
te...@cs.weber.edu
---
Any opinions in this posting are my own and not those of my present
or previous employers.
======== (1) ================================================================
Yup, you got the right place. I have ported Motif to 386BSD and Linux
and am about to do the same with NetBSD. I'm appending my standard
blurb; if you have other questions just email me.
Enrico
************************************************************************
* *
* OSF/Motif 1.1 for 386BSD-0.1 *
* *
* Ported by *
* Enrico Badella *
* Soft*Star s.r.l. *
* Via Camburzano 9 *
* 10143 Torino *
* ITALY *
* *
************************************************************************
OSF/Motif 1.1 is finally available for 386BSD-0.1
The distribution is BINARY ONLY and is NOT FREE because of OSF licensing
policy and royalty payment scheme.
I have organized the following distribution tree:
bin:
mwm* mxgdb* uil* xgen*
include:
Mrm/ Xm/ uil/
lib:
X11/ libMrm.a libUil.a libXm.a
lib/X11:
XKeysymDB app-defaults/ config.Motif/ system.mwmrc
demos:
Imakefile hellohebrew/ motifburger/ mre/ widgetView/
Makefile hellokanji/ motifburint/ periodic/ xmsamplers/
dogs/ hellomotif/ motifgif/ periodickanji/
fonts/ motifanim/ motifshell/ uilsymdump/
xgenDemos:
Example Xgen.intro dp.xbm mtn.xbm
README.EX address.list gnu.xbm
As you will probably have noted I have included all the demos released
by OSF and two non OSF applications. One is mxgdb, a OSF/Motif frontend
to gdb written by jtsi...@sprite.ma30.bull.com. The other is xgen,
a very usefull tool for creating graphical frontends to shell scripts
or other Unix programs; xgen was written by ku...@zorro.cecer.army.mil.
Everything is ready to run; you just have to load it on your 386bsd machine.
The obvious prerequisite is that you have a working X11 on your system.
The size of a compressed tar distribution is about 20M. No OSF documentation
is included due to it's size and to the fact that it is easier to
purchase the Prentice-Hall books than printing it. Anyway if you really
want the original OSF PostScript docs we can make arrangements for including
it.
At the moment I have used three mediums to distribute the product
1) QIC-24 tape
2) ftp
3) floppies (tar or DOS)
The first two (tape and ftp) are preferred due to the effort required
in producing the floppies. The ftp method is probably the fastest since
all I need is a temporary account on your machine in order to upload
96 files of 204800 bytes each. From my location to any machine with a IP
address and a reasonable connection the transfer typically takes from 30
minutes to an hour.
The cost of OSF/Motif for 386BSD is $150 + POSTAGE. In case you choose
a ftp distribution there is no postage, but I will charge you a fixed $15
to cover my phone costs. Preferred payment method is with Visa or Master
Cards.
The postage costs for tape or floppies depend greatly on the destination
and the speed required. Simple AIR MAIL to USA is about $10, express mail
to Germany is DM 32 and a comparable price to the rest of Europe.
I can offer email, fax, phone (the order represents preference) support
for initial installation problems. There shouldn't be any since I've
been using it for a while with 386BSD 0.1.40 kernel available on
agate.berkeley.edu. I didn't try the binary distribution of XFree86
but used MIT X11R5 patchlevel 17 source tree + XFree86.
If you have questions or are interested in purchasing OSF/Motif for 386BSD
please contact me at
Soft*Star s.r.l
Via Camburzano 9
10143 Torino
Italy
phone +39-11-746092
fax +39-11-746487
email e...@icnucevx.cnuce.cnr.it (preferred)
e...@sgi.iunet.it
Thank you for your patience.
************************************************************************
* All trademarks are of whoever owns them *
************************************************************************
======== (1) ================================================================
======== (2) ================================================================
I ftp'd what is claimed to be the first public domain version of Motif
from somewhere in austrailia. I dont have it online right now but I will
post the location of it as soon as I get some time as I am right in the middle
of finals. I hope to find time by friday. The name of the file was xmfm????
something. I cannot remember exactly. It is somewhere in .oz.
[ ... ]
BTW, If there is enough interest I could begin to maintain an archive here
on plains.nodak.edu. I am sure that this package will need some porting
to 386bsd. I was planning to do it over the holidays. Feed back is
welcome
--
Chris
tj...@plains.nodak.edu Christopher C. Tjon
uunet!plains!tjon (UUCP) Box 5224
tjon@plains (Bitnet) Fargo, ND 58105-5224
This message brought to you by the letters X and S and the number 3.
======== (2) ================================================================
======== (3) ================================================================
-------------------------------------------------------------------------
* OSF/Motif 1.1.2 for 386BSD-0.1 *
* *
* Ported by *
* Enrico Badella *
* Soft*Star s.r.l. *
* Via Camburzano 9 *
* 10143 Torino *
* ITALY *
OSF/Motif 1.1.2 is finally available for 386BSD-0.1
The distribution is BINARY ONLY and is NOT FREE because of OSF licensing
policy and royalty payment scheme.
The cost of OSF/Motif for 386BSD is $120 + postage. This is the lowest
fee I could come up with to cover all the expenses mainly due to
royalties to OSF and cartridge tapes.
I have organized the following distribution tree:
bin:
mwm* mxgdb* uil* xgen*
include:
Mrm/ Xm/ uil/
lib:
X11/ libMrm.a libUil.a libXm.a
lib/X11:
XKeysymDB app-defaults/ config.Motif/ system.mwmrc
demos:
Imakefile hellohebrew/ motifburger/ mre/ widgetView/
Makefile hellokanji/ motifburint/ periodic/ xmsamplers/
dogs/ hellomotif/ motifgif/ periodickanji/
fonts/ motifanim/ motifshell/ uilsymdump/
xgenDemos:
Example Xgen.intro dp.xbm mtn.xbm
README.EX address.list gnu.xbm
The size of the distribution is a little over 18M.
As you will probably have noted I have included all the demos released
by OSF and two non OSF applications. One is mxgdb, a OSF/Motif frontend
to gdb written by jtsi...@sprite.ma30.bull.com. The other is xgen,
a very usefull tool for creating graphical frontends to shell scripts
or other Unix programs; xgen was written by ku...@zorro.cecer.army.mil.
Everything is ready to run; you just have to extract the tape in
your X11 installation.
I can offer email, fax, phone (the order represents preference) support
for initial installation problems. There shouldn't be any since I've
been using it for a while with 386BSD 0.1.40 kernel available on
agate.berkeley.edu. I didn't try the binary distribution of XFree86
but used MIT X11R5 patchlevel 17 source tree + XFree86.
If you have questions or are interested in purchasing OSF/Motif for 386BSD
please contact me at
Soft*Star s.r.l
Via Camburzano 9
10143 Torino
Italy
phone +39-11-746092
fax +39-11-746487
email e...@icnucevx.cnuce.cnr.it (preferred)
e...@felix.sublink.org
Just a last note. Please don't start a flame war about commercial use
of usenet. I don't want (and won't) become rich with this operation, my main
activity is completely different. I'm just very please with 386BSD
(it's great!) and it would be nice to turn it in a new ODT; I'll soon
start investigating the portability of Postgres or OBST on 386BSD.
Thank you for your patience.
--
Enrico Badella Internet: e...@felix.sublink.org
Soft*Star s.r.l. e...@icnucevx.cnuce.cnr.it
Via Camburzano 9 Phone: +39-11-746092
10143 Torino, ITALY Fax: +39-11-746487
======== (3) ================================================================
======== (4) ================================================================
What I am about to describe is the terms surrounding Motif 1.2; we don't
license 1.1 any more.
There are, broadly, two kinds of license from us: the Limited Distribution
Right, and the Full Distribution Right (there are variants on the latter
for large users). These have a license fee, of $2k and $15k respectively,
and get you source.
Second, there are royalties. These start at $40 and are payable to us when
you ship binaries (libraries, shared or not, and/or mwm) onto a given
system outside your company. Once anyone -- you or someone else -- has
done this, the system has become "licensed".
Shipping binaries outside can be done only by those who hold a FDR, or by
their sublicensees. Those who hold a LDR can install binaries within their
organization. Anyone with a legal binary (FDR, LDR, or the customer of a
FDR) or can ship runtimes -- statically linked, complete programs, or
programs that depend on pre-installed shared libraries -- without fee, but
*only* to a system that was licensed in the previous paragraph.
[ ... ]
The royalty has been $40 (going down to $10 in volume) since day 1. How
much you charge your customers for them is up to you.
--
David Brooks dbr...@osf.org
Open Software Foundation uunet!osf.org!dbrooks
121 more days of Bush Presidency!
======== (4) ================================================================
John Fanelli
In a recent Unigram X article discussing binary standards for Unix running on In
tel, they quoted SCO as having a 60% market share of Unix on Intel.
john fanelli
Stephen R. Savitzky
Perhaps somebody could convert these percentages to absolute numbers?
--
\ --Steve Savitzky-- \ 343 Leigh Ave \ REAL HACKERS USE AN AXE!
\ st...@crc.ricoh.COM \ San Jose, CA 95128 \ Free Cyberia!
\ w: 415-496-5710 \ h:408-294-6492 \
\_________________________________________________________________________
Larry D Snyder
>Well someone posted that for 1992 SCO sales revenue was $200,000,000
>So lets assume that each unix system cost $500 (way below wholesale
>price :-) )
>This means that SCO shipped approximately 400,000 unix systems for the
>year 1992.
I'll bet you a large portion of that income is from suckers who paid
$CO for "support" since it is no longer free with their products
--
Larry Snyder
la...@gator.oau.org
KZU...@lstc2vm.stortek.com
ke...@ksmith.com (Keith Smith) writes:
>
>In article <1993Dec10.0...@anomaly.sbs.com>,
>Rev. Michael P. Deignan <kd...@anomaly.sbs.com> wrote:
>>dcm...@ritvax.isc.rit.edu (Dan Mattrazzo) writes:
>>
>>> Yes, Linux comes with gcc, ++, mail, TCP/IP, CD-ROM support,
>>> soundboard support, XWindows, on and on. And it's free.
>>
>>Unfortuntely, it doesn't support a fraction of the device drivers
>>that are available on the market for other devices, such as a
>>Always Technology SCSI board, or a Specialix SIO card.
>>
>it user supported or SCO supported or Card vendor specific?
>
>BTW I'd lay odds that SCO Unix is supported by and supports more
>hardware than _ANY_ other Unix PERIOD, so any driver support arguments
>vs. SCO are a lose for the other side.
>
Not to mention the fact that there is support for the Always IN2000 now
available on sunsite and tsx-11, and has been for a few months. Besides
the fact that I wouldn't call the Allways card a standard to be focused on.
Not saying it a bad card, its just not the standard. There are some gaps in
Linux yes, but we got the nets best minds filling them in.
-------------------------------------------------------------------------
| L IIIII NN N U U X X Having Problems? Be sure to pick |
| L I N N N U U X X up the new book 'Linux Installation|
| L I N N N U U X and Getting Started' by Matt Welsh |
| L IIIII N NN UUU X X Available at tsx-11.mit.edu in |
| LLLLL X X /pub/linux/docs/LDP and other sites|
| -Kevin Zupan |
| Out to fix what Windoze MS'd up! Kzu...@lstc2vm.stortek.com |
-------------------------------------------------------------------------
Keith Smith
>
>> Yes, Linux comes with gcc, ++, mail, TCP/IP, CD-ROM support,
>> soundboard support, XWindows, on and on. And it's free.
>
>Unfortuntely, it doesn't support a fraction of the device drivers
>that are available on the market for other devices, such as a
>Always Technology SCSI board, or a Specialix SIO card.
>
Funny, I missed the soundblaster support in SCO. Is it on sosco? Is
it user supported or SCO supported or Card vendor specific?
BTW I'd lay odds that SCO Unix is supported by and supports more
hardware than _ANY_ other Unix PERIOD, so any driver support arguments
vs. SCO are a lose for the other side.
--
Keith Smith ke...@ksmith.com 5719 Archer Rd.
Digital Designs BBS 1-919-423-4216 Hope Mills, NC 28348-2201
Somewhere in the Styx of North Carolina ...
Bill Bogstad
Stuart Lynne <s...@vanbc.wimsey.com> wrote:
>...
>This is a fact that is *not* lost on application software vendors. People
>who don't like to pay money for the OS also don't usually like to spend
>money on application software. So the vendors don't worry about the number
>of users of Linux/Minix/*bsd/?? simply because they don't view them as
>potential customers.
Please. Nobody ever likes paying money for anything if they can get
the exact same thing for free. ("same thing" includes support, future
updates, compatibility with the rest of the world, etc.) Except for hacker
types (myself included), most computer users really don't care what operating
system they use AS LONG AS IT LETS THEM DO WHAT THEY WANT WITH THEIR
COMPUTER. The operating system doesn't do that for most of them and is
essentially worthless to those users. For most people what they want is word
processing, spreadsheets, and databases. A small, struggling company with
good desktop publishing software for Unix systems might consider some kind
of joint distribution with one of the even smaller companies that distributes
packaged Linux or *BSD systems. Get a couple of companies together with a
good product in each of the 3 catagories above and who knows what would
happen. Or maybe not...
Bill Bogstad
bog...@cs.jhu.edu
Amancio Hasty Jr
>In article <1993Dec11.0...@ksmith.com>
>ke...@ksmith.com (Keith Smith) writes:
>
>>
>>In article <1993Dec10.0...@anomaly.sbs.com>,
>>Rev. Michael P. Deignan <kd...@anomaly.sbs.com> wrote:
>>>dcm...@ritvax.isc.rit.edu (Dan Mattrazzo) writes:
>>>
>>>> Yes, Linux comes with gcc, ++, mail, TCP/IP, CD-ROM support,
>>>> soundboard support, XWindows, on and on. And it's free.
>>>
>>>Unfortuntely, it doesn't support a fraction of the device drivers
>>>that are available on the market for other devices, such as a
>>>Always Technology SCSI board, or a Specialix SIO card.
>>>
>>Funny, I missed the soundblaster support in SCO. Is it on sosco? Is
>>it user supported or SCO supported or Card vendor specific?
>>
>>BTW I'd lay odds that SCO Unix is supported by and supports more
>>hardware than _ANY_ other Unix PERIOD, so any driver support arguments
>>vs. SCO are a lose for the other side.
>>
>Not to mention the fact that there is support for the Always IN2000 now
>available on sunsite and tsx-11, and has been for a few months. Besides
>the fact that I wouldn't call the Allways card a standard to be focused on.
>Not saying it a bad card, its just not the standard. There are some gaps in
>Linux yes, but we got the nets best minds filling them in.
>
Once I had a nightmare , a salesman trying to sell me the SCO developement
system for $2000. However, *bsd and linux bail us out :-)
Granted the price for SCO may have dropped recently but it is too late
for developers like me.
Amancio Hasty Jr
>In article <hastyCH...@netcom.com> ha...@netcom.com (Amancio Hasty Jr) writes:
>>The FreeBSD or NetBSD distribution includes binaries and sources for
>>gcc/tcp/ip/NFS/X cdrom support. NetBSD offers the ability to run
>> BSDI binaries and has dynamic shared libraries also it is being
>> ported to different platforms.
>>
>>Oops I forgot not mentioned no motif.
>
>
>Actually, several Motif and Motif-like and Motif-clone implementations are
>available for one or both systems (4 actually, although one is a version
>1.1 and one is a version 1.1.2 from the same source). Information follows:
>
My point is that Motif is not available with any *bsd CDROM distribution
unless Enrico wants to start distributing FreeBSD and/or NetBSD
also instead of costing $40 it can raise the price to close to $200
for the CDROM distribution.
I never heard of this motif clone it merits further investigation.
Any way thanks for the update.
Amancio Hasty Jr
>In article <2ec989$k...@vanbc.wimsey.com>,
>Stuart Lynne <s...@vanbc.wimsey.com> wrote:
>>...
>>This is a fact that is *not* lost on application software vendors. People
>>who don't like to pay money for the OS also don't usually like to spend
>>money on application software. So the vendors don't worry about the number
>>of users of Linux/Minix/*bsd/?? simply because they don't view them as
>>potential customers.
>
My recipe :-)
Well there is university ingres database and a tcl interface for it.
As for a spreadsheet it shouldn't be too hard to come up with one since
we have tcl. We could for instance write an X app with an embeded tcl
procedure for handling the expressions in fact it would make a very cool
cool spread sheet.
A solid word processor is lacking unless someone wants to polish
interviews idoc and we could have several choices for paint programs
xfig or idraw.
Vadim Antonov
: o Support QIC-40/QIC-80 drivers unless they have rewritten
: clock.c for sufficient timer resoloution (commercial UNIX
: timer resoloution is 10ms, and this is INSUFFICIENT).
That's a piece of Jesus M's mythology. You don't need hi-res timer
to do QIC-117, period. The standard is obscure and it is not obvious
from the beginning.
: o Don't support mounting of ISO9960 CD ROMs unless they have
: installed the PD "File System Survival Kit".
BSD/386 has built-in 9960 FS.
: o Support sound cards (NeXTStep isn't UNIX, sorry).
BSD/386 has SoundBlaster drivers in standard distribution.
: o Support Mitsumi CD ROM drives.
The same.
: o Directly support EISA devices (example: SCSI DMA not using
: bounce buffers when more than 16M of memory exists).
The same.
: o Support WABI.
Solaris.
: o Support Intel video capture boards.
How many people have them to make development of the driver to pay
for itself?
: o Cost the same as (or less than) DOS+Windows.
Sue Microsoft for dumping. Two extra hundred bucks do not make much
difference when you're buying a $3000 P(iece of) C(rap) anyway.
--vadim
PS I do not work for BSD Inc or Sun Microsystems or their subsidiaries.
Amancio Hasty Jr
>In <2ebb20$q...@u.cc.utah.edu>, A Wizard of Earth C (te...@cs.weber.edu) wrote:
>
>: o Support sound cards (NeXTStep isn't UNIX, sorry).
>
>BSD/386 has SoundBlaster drivers in standard distribution.
The BSDI Sound driver was contributed by a netter.
Besides the SoundBlaster is pretty much obsolete by now.
The current high quality sound cards use wave table synthesis - like
the Gravis Ultra Sound (GUS) card capable of putting out dolby, 3d sound,
supports midi and it sounds very good. Sure the SBPRO can do midi
at the expense of a lot of cpu. The sdk for the GUS is freely available
on the net.
Linux already has support for the GUS and we are wrapping up the support
under NetBSD-current.
Would not be suprise if the ISO 9960 support was also contributed by
a netter.
Good Nite,
Alan Drew
>In article <1993Dec10.0...@anomaly.sbs.com> kd...@anomaly.sbs.com (Rev. Michael P. Deignan) writes:
>>dcm...@ritvax.isc.rit.edu (Dan Mattrazzo) writes:
>>
>>> Yes, Linux comes with gcc, ++, mail, TCP/IP, CD-ROM support,
>>> soundboard support, XWindows, on and on. And it's free.
>>
>>Unfortuntely, it doesn't support a fraction of the device drivers
>>that are available on the market for other devices, such as a
> Operating systems?--^^^^^^^
>>Always Technology SCSI board, or a Specialix SIO card.
>Are you referring to DOS device drivers (which don't belong in these
>news groups) or are you referring to other OS's having more device
>support?
>What UNIX systems *do* support the Specialix SIO card? (If it is a
>board without an on-board processer, I guarantee you one of Linux
>or *BSD supports it).
Is this a serious question? Do people *really* need me to answer this one?
I am continually amazed that people who can be so ignorant
post with such *authority*
Alan Drew
Speaking from Specialix, not necessarily on behalf of.
--
Larry D Snyder
>Once I had a nightmare , a salesman trying to sell me the SCO developement
>system for $2000. However, *bsd and linux bail us out :-)
>Granted the price for SCO may have dropped recently but it is too late
>for developers like me.
But they added additional restrictions as to the number of users for each
license -- which actually is still more the the majority of other Unix
products on the market
--
Larry Snyder
la...@gator.oau.org
Jonathan Badger
>>What UNIX systems *do* support the Specialix SIO card? (If it is a
>>board without an on-board processer, I guarantee you one of Linux
>>or *BSD supports it).
>Is this a serious question? Do people *really* need me to answer this one?
>I am continually amazed that people who can be so ignorant
>post with such *authority*
>Alan Drew
>Speaking from Specialix, not necessarily on behalf of.
And, I myself am amazed that companies think that their products are so
important that a system that doesn't support them is worthless. Rather to
the contrary, since Linux and *bsd drivers are written by people who need to
use the hardware, the presence or absense of a driver is a good litmus test
for whether a piece of hardware is considered useful...
Lawrence Kirby
>>What UNIX systems *do* support the Specialix SIO card? (If it is a
>>board without an on-board processer, I guarantee you one of Linux
>>or *BSD supports it).
>
>
>Is this a serious question? Do people *really* need me to answer this one?
having Linux drivers would certainly open up a large market for them. After
all anybody who hasn't spent lots of money on an operating system will have
to spend it on something else! :-)
-----------------------------------------
Lawrence Kirby | fr...@genesis.demon.co.uk
Wilts, England | 7073...@compuserve.com
-----------------------------------------
Fredrik Staxang)
>>>What UNIX systems *do* support the Specialix SIO card? (If it is a
>>>board without an on-board processer, I guarantee you one of Linux
>>>or *BSD supports it).
>>
>>
>>Is this a serious question? Do people *really* need me to answer this one?
Since the fellow from the manufacturer did not answer the question, I just add
what I think I know about the Specialix card. It is an intelligent
multiport serial card that, I believe, only buffers output when in raw mode.
That means it's fine for terminals but not so great for inbound file transfers.
--
Fredrik Stax\"ang |
CS Student@Uppsala University |
fs...@meryl.csd.uu.se |
(+46 18) 25 01 25 |
A Wizard of Earth C
inapropriate (I ought to add the UnixWare group, though...)
In article <2egnqh$q...@news.sprintlink.net> a...@sprintlink.net (Vadim Antonov) writes:
>In <2ebb20$q...@u.cc.utah.edu>, A Wizard of Earth C (te...@cs.weber.edu) wrote:
>: o Support QIC-40/QIC-80 drivers unless they have rewritten
>: clock.c for sufficient timer resoloution (commercial UNIX
>: timer resoloution is 10ms, and this is INSUFFICIENT).
>
>That's a piece of Jesus M's mythology. You don't need hi-res timer
>to do QIC-117, period. The standard is obscure and it is not obvious
>from the beginning.
Jesus M did spew an awful lot of mythology, but the timer resoloution
is a factor (though not in the way he thought it was). There are some
tight timing windows whose traditional soloutions have been driver
buzzloops (like DELAY() macros in *BSD).
Contrary to popular belief, the clock code really does want to be
higher resoloution than a proces quantum. I have been inside several
commercial QIC-40/80 drivers, and the *BSD one, and there is much
too much busy waiting.
This can be alleviated somewhat by putting a 32K transfer buffer in the
kernel for the driver, but it has to be contiguous, and the malloc
routines in *BSD don't like to do that without a bit of effort (hence
the 16K limit on the SB DMA that was recently announced pending a
clean way to pin the pages). There is still a tight window where
a 250uS timer is needed during reads unless you buzz down on the FDC
chip, and a not so tight timing constraint at 2-3mS (which is where the
buzz lopp really kills you).
You can't *not* staty in the kernel because you can't afford to not
be called in time -- you end up having to resync the tape, and then
nothing streams.
A Wizard of Earth C
>te...@cs.weber.edu (A Wizard of Earth C) writes:
>
>>What UNIX systems *do* support the Specialix SIO card? (If it is a
>>board without an on-board processer, I guarantee you one of Linux
>>or *BSD supports it).
>
>Is this a serious question? Do people *really* need me to answer this one?
Actually it was, and it was answered already in private email by
c...@specialix.co.uk. I would be happier if he had included board
info so I could tell how hard it would be to support a PD driver
for the thing, or if your company actually offered a source
distributable driver that could be adapted (it was also a sideways
question as to whether "it" was multiple boards or a single board
type, and whether or not "it" was a "dumb" board or had a processer).
I will try to be more specific next time, now that I know you are
watching over my shoulder.
>I am continually amazed that people who can be so ignorant
>post with such *authority*
I don't see how a question of this type purports to convey authority,
nor what it purports authority on.
Alan Drew
>c...@specialix.co.uk (Alan Drew) writes:
>>>What UNIX systems *do* support the Specialix SIO card? (If it is a
>>>board without an on-board processer, I guarantee you one of Linux
>>>or *BSD supports it).
>>Is this a serious question? Do people *really* need me to answer this one?
>And, I myself am amazed that companies think that their products are so
>important that a system that doesn't support them is worthless. Rather to
>the contrary, since Linux and *bsd drivers are written by people who need to
>use the hardware, the presence or absense of a driver is a good litmus test
>for whether a piece of hardware is considered useful...
Read the post slightly more carefully. I never stated anywhere that I consider
Linux or BSD to be worthless. All I did was ask a question, some dumb hick
fool asks which UNIX systems *do* support Specialix cards? That is a stupid
question asked by someone obviously showing his ignorance of the marketplace
in question.
As to the worthlessnes of linux and BSD or for that matter *ANY* operating
system unix or otherwise, as a company we are continually watching the
marketplace and evaluating where we need to port drivers for next.
Also *I* have been tasked with finding somebody somewhere who is capable of
porting our drivers to Linux, during that proces I have also been put in
touch with somebody who would like to port them to BSD aswell, *THAT* is how
worthless we see linux and BSD.
It is not that we consider our products so important that we ignore a system
that doesn't support them. The fact is that *NO* operating system by default
will support a Specialix Intelligent IO controller, we have to do it ourselves
with or without the help of the OS vendor.
That fact is true of *ANY* operating system and *ANY* intelligent IO
controller (to the best of my knowledge).
Alan Drew
Posting from Specialix, not necessarily on behalf of.
--
Alan Drew
>...and, you forget, a litmus test for whether the programming information is
>available or can be reconstructed to some extent. Consider the various hacks
>for Diamond cards, for example. If the Specialix SIO programming information
>isn't readily available and can't easily or safely be inferred by poking about
The information can be made available on application, there are reasons
why we don't dump our source code on ftp sites. On the other hand if there
is a reason why you would like a copy of our DTK, then *ASK* if it's valid
chances are you'll get one.
more info from alan.drew or cdh @specialix.co.uk or @specialix.com
they all get to me.
Alan
Speaking from Specialix, not necessarily on behalf of.
--
Alan Drew
>>In article <cdh.75...@specialix.co.uk> c...@specialix.co.uk writes:
>>>>What UNIX systems *do* support the Specialix SIO card? (If it is a
>>>>board without an on-board processer, I guarantee you one of Linux
>>>>or *BSD supports it).
>>>
>>>
>>>Is this a serious question? Do people *really* need me to answer this one?
>Since the fellow from the manufacturer did not answer the question, I just add
Well he didn't ask what an SI card did......., he just asked what unix systems
*do* support an SI card, which I consider to be a very silly question.
>what I think I know about the Specialix card. It is an intelligent
>multiport serial card that
The above stamenet is true......
> I believe, only buffers output when in raw mode.
>That means it's fine for terminals but not so great for inbound file transfers.
Err do you mean if you are in raw mode then output only is buffered (i.e not
input), or do you mean that if you are not in raw mode than no output
bufering is done?
Either way its not true, input or ouput data gets buffered on the host
regardless of raw mode or not.
As for being OK for terminals and not file transfer. Ever seen a SLIP
line running at 57.6Kbaud? ftp over a slip line at 57.6Kbaud works okay for
me thanks
Alan Drew
Speaking from not necessarily on behalf of Specialix.
--
Nigel Whitfield
>good desktop publishing software for Unix systems might consider some kind
>of joint distribution with one of the even smaller companies that distributes
>packaged Linux or *BSD systems. Get a couple of companies together with a
>good product in each of the 3 catagories above and who knows what would
>happen. Or maybe not...
Well, as the editor of a computer magazine, I wouldn't begin to
consider producing the magazine on a non-commercial platform. There is
great free software out there, but if anything goes wrong at a
critical moment, I have to know that I can ring someone up and get
them to sort it out.
Ultimately, in a situation like that, the economics work in favour of
the commercial solutions.
Nigel.
tim werner
>Well, as the editor of a computer magazine, I wouldn't begin to
>consider producing the magazine on a non-commercial platform. There is
>great free software out there, but if anything goes wrong at a
>critical moment, I have to know that I can ring someone up and get
>them to sort it out.
You mean ring them up and get assigned a problem report #.
Support is a good thing when they already know about your problem and
can give you a work-around and send you a patch tape or new rev, but
it doesn't always work that way. Sometimes they tell you they will have
to pass the problem to one of the 'experts', and call you back later.
Sometimes you call support and the only thing that happens is that you
end up explaining to some dummy how the software is supposed to work. :-(
Sometimes they never solve the problem. Just because they are charging
you money to hold your hand doesn't make them any smarter than you.
Or does it? :-)
At least when you have the source there's a chance you, or someone on
the net, can figure it out.
Thor Lancelot Simon
Unless one accepts the (IMHO ludicrous) assertion that a commercial OS vendor
would within the space of time presented by a "critical moment" fix its OS
were that broken (as commerical OSes clearly are just as often as free ones,
if not more), you whole article is logically inconsistent.
--
Thor Lancelot Simon t...@panix.COM
But as he knew no bad language, he had called him all the names of common
objects that he could think of, and had screamed: "You lamp! You towel! You
plate!" and so on. --Sigmund Freud
Stuart Lynne
>were that broken (as commerical OSes clearly are just as often as free ones,
That's a pretty sweeping statement.
Can you please supply some quantifiable data, the metric's you're using,
the method used to collect same etc.
Or is this just a subjective, personal opinion? In which case some
background on how you come to this conclusion would help us all evaluate
whether we agree with it.
--
Stuart Lynne <s...@wimsey.com> ........... 604-936-8649(voice) 604-937-7718(fax)
Installable Software for SCO UNIX ............ ftp.wimsey.com:~ftp/pub/wimseypd
Internet SCO UNIX troubleshooting ......................UNIX Facsimile Software
Rev. Michael P. Deignan
>Funny, I missed the soundblaster support in SCO. Is it on sosco? Is
>it user supported or SCO supported or Card vendor specific?
Soundblaster? What's that? Oh, you mean one of those little kiddie toy
things with a joystick port that makes k-k00l "bang-bang" sounds when
you press the joystick button? No, I haven't seen support for that in
SCO software... Then again, SCO UNIX isn't a kiddie-game OS. If you
want that, go run Messy-DOS.
MD
--
-- Michael P. Deignan, KD1HZ -
-- Internet: kd...@anomaly.sbs.com - I never tell the truth, because I
-- UUCP: ...!uunet!anomaly!kd1hz - I don't believe that there is such
-- AT&TNet: 401-273-4669 - a thing...
Mark A. Davis
>Or does it? :-)
There is a line of responsibility and exchange which does not exist with
"free" software. This line is very important to businesses which use
computers for real work. Not all commercial software packages have great
support, but there is still some line of responsibility; since their sales
are linked to word-of-mouth recommendations and such.
>At least when you have the source there's a chance you, or someone on
>the net, can figure it out.
Not really. That is mostly a comparison of small "free" software sources
compared to very large commercial sources. Besides, there is much more
chance that nothing will happen when you post a plea for help on the net
about "free" software..... there is no real reason why someone must or
should help out. AND it works both ways- people also find support for
commercial software on the net too.... no code hacking, but experiences,
configuration, application, troubleshooting, update info, etc....
--
/--------------------------------------------------------------------------\
| Mark A. Davis | Lake Taylor Hospital | Norfolk, VA (804)-461-5001x431 |
| Sys.Administrator| Computer Services | ma...@taylor.wyvern.com .uucp |
\--------------------------------------------------------------------------/
Amancio Hasty Jr
We no longer support that version; better yet we no longer support
that product. Or, how about 4 week turn around to just tell you
that they receive your problem statement and yes they have given it
a lot of thought and will in the future get back to you.
Does it ring a bell :-)
Amancio Hasty Jr
>ke...@ksmith.com (Keith Smith) writes:
>
>>Funny, I missed the soundblaster support in SCO. Is it on sosco? Is
>>it user supported or SCO supported or Card vendor specific?
>
>Soundblaster? What's that? Oh, you mean one of those little kiddie toy
>things with a joystick port that makes k-k00l "bang-bang" sounds when
>you press the joystick button? No, I haven't seen support for that in
>SCO software... Then again, SCO UNIX isn't a kiddie-game OS. If you
>want that, go run Messy-DOS.
It is true that SBPRO generates kiddie toy type sound.
However, I have my Gravis Ultra Sound connected to my $4000 stereo
generating sweet music with some pieces the GUS puts out dolby
sound. The sound driver is available for linux, freebsd, and netbsd.
Is nice to kick back and play a cool midi file in the background.
But then again you can't really do that under SCO so it will
be very difficult for you to realize.
Mark A. Davis
>In article <1993Dec15....@anomaly.sbs.com> kd...@anomaly.sbs.com (Rev. Michael P. Deignan) writes:
>>
>>Soundblaster? What's that? Oh, you mean one of those little kiddie toy
>It is true that SBPRO generates kiddie toy type sound.
>However, I have my Gravis Ultra Sound connected to my $4000 stereo
>generating sweet music with some pieces the GUS puts out dolby
>sound. The sound driver is available for linux, freebsd, and netbsd.
>Is nice to kick back and play a cool midi file in the background.
>But then again you can't really do that under SCO so it will
>be very difficult for you to realize.
Please people, let's get real.... if there were a demand for it on SCO, then
it would be there either because paying SCO customers foster a response in
the free market or because the "free" (PD) domain jumps on it.
(Of course there are often demands for things which never seem to happen).
The above statement could easily be turned around to say "I do voice mail
on SCO through a commercial package.... I don't see LINUX or freeBSD
doing that" (or countless other examples, how about multiprocessing, etc)
A Wizard of Earth C
>things with a joystick port that makes k-k00l "bang-bang" sounds when
>you press the joystick button? No, I haven't seen support for that in
>SCO software... Then again, SCO UNIX isn't a kiddie-game OS. If you
>want that, go run Messy-DOS.
Is that what you use your soundblaster for?
I tend to used them in only two minor places:
(1) MIME encapsulated voice mail attachements to email.
(2) PBX voice mail services.
I use them this way under UNIX, BTW.
Sun machines come with microphones now to allow you to use the included
email-with-integrated-voice-mail that comes on the machine.
Gee it would be nice if I could send these documents to SCO systems
too and have them be usable.
Gee it would be nice if I could run PBX software on an SCO machine, but
I can't.
A Wizard of Earth C
>>were that broken (as commerical OSes clearly are just as often as free ones,
>
>That's a pretty sweeping statement.
>
>Can you please supply some quantifiable data, the metric's you're using,
>the method used to collect same etc.
I'm not the initial poster, but:
>Internet SCO UNIX troubleshooting ......................UNIX Facsimile Software
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
Surely as an Internet SCO UNIX troubleshooter, you are aware of the Lachman
bug where the remote diconeect removes the local context on the first error
without attempting to recover, such that (for both SCO UNIX and Xenix):
/* fd = socket*/
write( fd, buf, len); /* not checked -- yields -1*/
if ( rlen = read( fd, buf, len)) == -1) { /* EXITS!*/
...
(1) The remote side closes the fd, the local write fails with -1
and the streams context for the fd is destroyed.
(2) A subsequent second operation (in this case a read) on the
invalid fd does not return an error (per the documentation);
instead the process is kicked as if it called exit(2).
I *should* be able to deal with the error on the read *by itself*...
I don't care how many errors I get, until I close(2) it, it has no
right to destroy the context (PS: this was reported in 1989).
Or the select() (to narrow ourselves down to Xenix only) does not
work on streams fd's -- poll must be used instead (hence the hack
function XSelect() in the X lib).
Or the select() (to narrow ourselves to SCO UNIX only) does not
have sufficient resoloution (being limited to the clock update
frequency instead of the clock frequency) limiting it to 10ms
resoloution (being implemented in the library using poll doesn't
help, either), in clear violation of SVID Third Edition (not to
mention getitimer/setitimer (in the RT section)).
There are bunches more examples of "well known problems" besides these
(Check out Eric Raymonds PC UNIX FAQ) that have yet to be corrected.
Something being commercial no more enobles it than it being non-commercial;
broke code is broke code.
Dan Rosenblatt
|> ke...@ksmith.com (Keith Smith) writes:
|>
|> >Funny, I missed the soundblaster support in SCO. Is it on sosco? Is
|> >it user supported or SCO supported or Card vendor specific?
|>
|> Soundblaster? What's that? Oh, you mean one of those little kiddie toy
|> things with a joystick port that makes k-k00l "bang-bang" sounds when
|> you press the joystick button? No, I haven't seen support for that in
|> SCO software... Then again, SCO UNIX isn't a kiddie-game OS. If you
|> want that, go run Messy-DOS.
Yes, SoundBlaster support is available through sosco.sco.com.
Watch out, Merge (the DOS emulator) doesn't fully support the
SoundBlaster because there is some sort of DMA limitation/bug
that will be fixed in the next major release of Merge from Locus.
Dan R
Stuart Lynne
A Wizard of Earth C <te...@cs.weber.edu> wrote:
>In article <2em4ds$n...@vanbc.wimsey.com> s...@vanbc.wimsey.com (Stuart Lynne) writes:
>>>were that broken (as commerical OSes clearly are just as often as free ones,
>>
>>That's a pretty sweeping statement.
>>
>>Can you please supply some quantifiable data, the metric's you're using,
>>the method used to collect same etc.
>
>bug where the remote diconeect removes the local context on the first error
>without attempting to recover, such that (for both SCO UNIX and Xenix):
What does this have to do with the statement "as commercial OSes clearly are
{broken} just as often as free ones." ?
I certainly never maintained that commercial OSes are bug free. And
certinaly looking at the number of patches available for things like Solaris
or SCO we know for certain that they are not. My only beef is that this
seems to be a broad and sweeping statement that is NOT backed up by any
facts. And seems to be a subjective personal opinion. Which is ok, most of
the net is just that.
My personal opinion is that with a large piece of software (such as an OS)
the longer it has been sold, the larger the user base, the lower the number
"serious" bugs (i.e. showstoppers, prevent the system from being useful,
panic etc type bugs). More users equals more bugs found, more time equals
more bugs fixed.
If we can agree that (for example) there are more users of SCO or Sun OS or
Sparc than Linux or netBSD or BSD386. And that they have been in production
mode for longer. So I would contend that they probably have less problems
with them. Of course this is a personal opinion. I don't work for any of the
above so I don't have access to their buglists, or customer counts.
>There are bunches more examples of "well known problems" besides these
>(Check out Eric Raymonds PC UNIX FAQ) that have yet to be corrected.
>
>Something being commercial no more enobles it than it being non-commercial;
>broke code is broke code.
Sure. But do you *seriously* believe that every user of SCO or Xenix or
Solaris or (pick your favorite UNIX version) could install a PD UNIX
tomorrow at 9:00AM and have everything up and running by suppertime? We're
talking about a lot of users who barely know how to log in once they get it
installed.
I know *I* could install a PD Unix here and have lot's of fun. But I don't
think it would be usable on our main machine. We'd have to get drivers for
Digiboard and Adaptec 2742's. And DAT's, etc. Not too mention I'd have to
recompile everything from scratch that we've worked on for the last five
years under SCO. And implement a new set of sysadmin tools. And make sure
it's secure. And of course once I'd done all that I'd be an netBSD sysadmin
expert and I could probably make lot's of money locally doing that. Not!
Anyhow, have a nice day. There is certainly room in the market for both
commercial and free OS's. I don't think they really compete for the same
users at this point.
--
Stuart Lynne <s...@wimsey.com> ........... 604-936-8649(voice) 604-937-7718(fax)
Installable Software for SCO UNIX ............ ftp.wimsey.com:~ftp/pub/wimseypd
Stuart Lynne
>>Soundblaster? What's that? Oh, you mean one of those little kiddie toy
>
>
>I tend to used them in only two minor places:
>
>(1) MIME encapsulated voice mail attachements to email.
>
>(2) PBX voice mail services.
>
>I use them this way under UNIX, BTW.
>
>Sun machines come with microphones now to allow you to use the included
>email-with-integrated-voice-mail that comes on the machine.
>
>Gee it would be nice if I could send these documents to SCO systems
>too and have them be usable.
>
>Gee it would be nice if I could run PBX software on an SCO machine, but
>I can't.
Sure you can. Just buy dialogic or brooktrout boards which are designed to
implement voice response and fax systems. Trying to hack a VRS from a sound
card is just not the way to go when you want to support a dozen or so lines.
Those solutions have been available for about five years. Pricey, but they
do work well.
The sound blaster would be good to implement the same types of things we're
now seeing on workstations like Sun's and Mac's. With a microphone and
playback for voice mail etc.
But as much as SCO doesn't like to admit it, SCO UNIX is really best in
supporting large numbers of users on terminals (now changing to X Terminals)
running non workstation type applications.
But it certainly would not be hard to do what you want under SCO ODT with
appropriate drivers. And when the demand is there I'm sure that someone will
provide appropriate drivers. If there is a driver available for netBSD etc
it wouldn't take very long to get ported to SCO :-)
I'm just about to buy a soundcard for the kid's. Anyone got a recommendation?
We feed Gravis news and mail. Maybe I should give them a call.
Amancio Hasty Jr
>ha...@netcom.com (Amancio Hasty Jr) writes:
>
>>In article <1993Dec15....@anomaly.sbs.com> kd...@anomaly.sbs.com (Rev. Michael P. Deignan) writes:
>>>
>>>Soundblaster? What's that? Oh, you mean one of those little kiddie toy
>
>>It is true that SBPRO generates kiddie toy type sound.
>>However, I have my Gravis Ultra Sound connected to my $4000 stereo
>>generating sweet music with some pieces the GUS puts out dolby
>Please people, let's get real.... if there were a demand for it on SCO, then
>it would be there either because paying SCO customers foster a response in
>the free market or because the "free" (PD) domain jumps on it.
>(Of course there are often demands for things which never seem to happen).
>
>The above statement could easily be turned around to say "I do voice mail
>on SCO through a commercial package.... I don't see LINUX or freeBSD
>doing that" (or countless other examples, how about multiprocessing, etc)
>
A reality check point here:
The GUS is ideal for the PD unix systems because we don't have
backwards compatibility issue like with the GUS and SBPRO utilities
under DOS;additionally, the GUS' SDK is freely available so all we
have to do is download the SDK from some ftp site.
The GUS is ideal for the home use because I can see a need to put
out high quality sound. To be honest, I don't see much of a need
for high quality voice mail. For voice mail all is required is
a maximum sampling rate of 12k, unless you want to mail a song :-)
If SCO's market is content with mediocre sound does not mean
that we have to follow suit.
As for multimedia capability for the free unix systems, there
is safe-tcl:
======================================
ANNOUNCEMENT
We am pleased to announce an implementation of Enabled Mail (EM)
for many UNIX systems. This is a beta release.
The idea behind Enabled Mail is that messages contain programs
which get evaluated during delivery, receipt, and displaying. For
example, every time you receive a message, a program you specify
examines the message and performs some actions, such as filing,
sending a note to your pager, etc. This is an example of
receipt-time EM. Alternately, you might send a program to
someone with the intent of having it execute when the show the
message. This is an example of activation-time EM.
Earlier systems have done bits and pieces of this, within the
context of a very specific environment. With EM, we have tried to
provide a general model which can be used in a variety of
environments.
The choice of our programming language is Tcl - Ousterhout's Tool
Command Language, which is rapidly becoming a popular systems
language. With Tcl, we have integrated support for
- MIME, so you can deal with multimedia messages;
- display environments, so you use different UI paradigms
(e.g., screen-based, window-based) depending on what the
recipient has; and,
- execution safety, so you don't have to worry about someone
sending you a malicious program.
In the beta period, we hope to gain experience both with the
technology choices we've made (e.g., Tcl) and portability of the
implementation we provide. Of course, we also hope that others
will implement EM for their (non-UNIX) systems.
======================================
Basically, the free unix systems are getting stronger literally
day by day.
News admin
Va negvpyr <CI3op5.1pq@da_vinci.it.uswc.uswest.com>
dxr...@newsit.mtt.it.uswc.uswest.com (Dan Rosenblatt) jevgrf:
} In article <1993Dec15....@anomaly.sbs.com>,
} kd...@anomaly.sbs.com (Rev. Michael P. Deignan) writes:
} |> ke...@ksmith.com (Keith Smith) writes:
} |>
} |> >Funny, I missed the soundblaster support in SCO. Is it on sosco? Is
} |> >it user supported or SCO supported or Card vendor specific?
} |>
} |> Soundblaster? What's that? Oh, you mean one of those little kiddie toy
The files are in the TLS section:
tls023.tar.Z Soundblaster driver
tls023d.tar.Z Still more improved driver (binaries only)
tls023.ltr cover letter and README
Available via both anonymous uucp and ftp.
SCO porters like little toys too. The official and announced support for
multimedia (whether de facto or industry standards) should not be that far
behind (I should think).
In any case, I think this thread's probably gotten way out of hand by now.
Please note the followups.
Many happy returns,
-timr
--
Tomorrow, and tomorrow, and tomorrow,
Creeps in this petty pace from day to day,
To the last syllable of recorded time;
And all of our yesterdays have lighted fools
The way to dusty death. Out, out, brief candle!
Life's but a walking shadow, a poor player
That struts and frets his hour upon the stage,
And then is heard no more; it is a tale
Told by and idiot, full of sound and fury,
Signifying nothing.
--Shakespeare, _Macbeth_
Nigel Whitfield
>
>At least when you have the source there's a chance you, or someone on
>the net, can figure it out.
I'm the editor of a magazine. Not a C programmer. Whatever platform
we're using, I want the operating system to work and not fall over at
a time when a couple of hours delay will cost us thousands of pounds.
The economics of the business demand that you have come-back on
support.
Nigel.
Nigel Whitfield
>
>Unless one accepts the (IMHO ludicrous) assertion that a commercial OS vendor
>would within the space of time presented by a "critical moment" fix its OS
>were that broken (as commerical OSes clearly are just as often as free ones,
>if not more), you whole article is logically inconsistent.
I can only assume that you don't know the first thing about magazine
publishing. Your sweeping statement seems, IMHO, typical of people who
want businesses to run unsupported software. I'm quite happy to do so
at home, where a short delay won't cost me lots of money.
When it comes to work, I have to know that there is support there that
is unconditional. It doesn't have to depend on my staff knowing how to
access the net, and the right person reading the right newsgroup at
the appropriate moment.
I can't afford to use a non-standard solution. I can't risk the
magazine not appearing because there's no one who can give me a level
of support that I'm happy to pay for.
There is a place for free software, and the price of much commercial
stuff is too high. I've even written editorials about it. But I know
the financial and personnel facts of the environment in which I work.
And running systems on an unsupported OS is not an option.
Nigel.
tim werner
>In article <1993Dec15.0...@news.csuohio.edu> thx...@babbage.cba.csuohio.edu (tim werner) writes:
>>
>>At least when you have the source there's a chance you, or someone on
>>the net, can figure it out.
>
>I'm the editor of a magazine. Not a C programmer. Whatever platform
>we're using, I want the operating system to work and not fall over at
>a time when a couple of hours delay will cost us thousands of pounds.
I can only repeat that paying someone for a support contract does not
guarantee that they will solve your problem when you call them.
I'm not saying that support is a waste of money, just that it's no
guarantee. Some companies have better support than others, but they
all have human beings working there. Just because you are paying them
doesn't mean Superman will answer the phone and come flying over to
save you. We have had problems that simply *never* got fixed.
A Wizard of Earth C
[ ... someone said "commerical OSes are broken just as often as free ones ... ]
>>>That's a pretty sweeping statement.
>>>
>>>Can you please supply some quantifiable data, the metric's you're using,
>>>the method used to collect same etc.
[ ... discussion of how one commercial OS is broken given as an example ... ]
>What does this have to do with the statement "as commercial OSes clearly are
>{broken} just as often as free ones." ?
You're probably right -- we can draw no conclusion; I didn't read this as
a request for metrics. But by the same token you can not present the
conclusion that "commerical OSes are *NOT* broken just as often as free
ones" without the same metrics
>My personal opinion is that with a large piece of software (such as an OS)
>the longer it has been sold, the larger the user base, the lower the number
>"serious" bugs (i.e. showstoppers, prevent the system from being useful,
>panic etc type bugs). More users equals more bugs found, more time equals
>more bugs fixed.
This is opinion that many people disagree with. I happen to be of the
"higher availability of source code equals more bugs fixed" religion
myself, and having occasion to work within an organization where a
blatant bug can't be fixed even if it is a simple fix without following
a week long (or longer) procedure, I can say that the find/fix cycle
seems a lot faster in the free OS's.
>If we can agree that (for example) there are more users of SCO or Sun OS or
>Sparc than Linux or netBSD or BSD386. And that they have been in production
>mode for longer. So I would contend that they probably have less problems
>with them. Of course this is a personal opinion. I don't work for any of the
>above so I don't have access to their buglists, or customer counts.
I agree that there are more commercial OS installations then free OS
installations, and I agree with them having been around longer, but I
disagree with your major premise that "More users equals more bugs found,
more time equals more bugs fixed", or even that if it were acceptable,
that commercial OS's didn't have more bugs to start with.
In my opinion, very little innovation and research really occur in
commercial products because innovation and research do not impact the
bottom line. Similarly, a problem that one user has, or for which a
workaround exists is unlikely to be fixed in a commercial environment
-- a commercial OS.
I don't see very many commercial products which have taken advantage of
Heidemann or Rosenthal's statckable file system technology, for example.
The vast majority of the commercial work that is being done is DOS and
Windows centric. It has only take 12 years for Microsoft to be on the
verge (with Chicago) of bumping up the 8.3 filename limit.
Also in my opinion, commercial OS implementations give the users what
they need to do the work they need done -- and stop there, at the bare
minimum requirements. Nothing else impacts the bottom line. What
else can explain the "File system Survival Kit" for SVR4 mounting of
DOS and ISO (CDROM) file systems having come out of modified code
from the *BSD projects instead of out of the commercial UNIX vendors?
The added convenince was outside the scope of the minimal feature set
and so was ignored as not impacting the bottom line. This has GOT to
change!
>>Something being commercial no more enobles it than it being non-commercial;
>>broke code is broke code.
>
>Sure. But do you *seriously* believe that every user of SCO or Xenix or
>Solaris or (pick your favorite UNIX version) could install a PD UNIX
>tomorrow at 9:00AM and have everything up and running by suppertime? We're
>talking about a lot of users who barely know how to log in once they get it
>installed.
We are talking release engineering problems; I believe several of the
free UNIX clones are extremely close to this goal; the "Slackware" Linux
distribution is arguably closest, with the majority of the people in
that arena having come from the DOS area. But *don't* confuse a release
engineering with stability. Many venders ship operating systems pre
installed, and most SCO installations are done by VARs or VADs ...NOT
by end users.
Your argument falls flat, because I contend that very few of the "every
user of SCO or Xenix or Solaris or (pick your favorite UNIX version)"
group given could install "SCO or Xenix or Solaris or (pick your
favorite UNIX version)" under the same criteria -- starting "tomorrow at
9:00AM and have everything up and running by suppertime".
>I know *I* could install a PD Unix here and have lot's of fun. But I don't
>think it would be usable on our main machine. We'd have to get drivers for
>Digiboard and Adaptec 2742's. And DAT's, etc. Not too mention I'd have to
>recompile everything from scratch that we've worked on for the last five
>years under SCO. And implement a new set of sysadmin tools. And make sure
>it's secure. And of course once I'd done all that I'd be an netBSD sysadmin
>expert and I could probably make lot's of money locally doing that. Not!
The Digiboard and Adaptec 2742 drivers might be a problem; if the Digiboard
is a "dumb" model, then it'll work -- the AHA2742 controllers will be a
problem because Adaptec moved their microcode to user space and then
made licensing decisions that make a distributable source implementation
a matter of rewriting the microcode for the sequencer (an AIC-7770).
There is some indication that the 1742 put less of a load on the host in
any case... with a UNIX implementation, you want the main processer
offloaded anyway, so a trade in for the 1742 controllers instead may even
give you a performance boost, even under SCO.
*BSD supports DAT drives.
*BSD supports "execution classes" in Alpha code, and there is a non-
distributable Xenix loader module (it'd have to be rewritten for you
to use it) that is about 1500 lines of code, so your recompilation
issues may be about to go away -- you'd just run your SCO binaries.
The sysadmin tools available depend on what operations you think you'll
typically perform; I was aware of a project to make POSIX compliant
sysadmin tools, but I don't know where that has gotten to.
The standard security tools from the net should run without modification.
Regards,
Mark A. Davis
>Is that what you use your soundblaster for?
>I tend to used them in only two minor places:
>(1) MIME encapsulated voice mail attachements to email.
>(2) PBX voice mail services.
>I use them this way under UNIX, BTW.
>Sun machines come with microphones now to allow you to use the included
>email-with-integrated-voice-mail that comes on the machine.
>Gee it would be nice if I could send these documents to SCO systems
>too and have them be usable.
You can, many Xterminals now include sound record and playback......
>Gee it would be nice if I could run PBX software on an SCO machine, but
>I can't.
Yes you can, there are commercial packages available for that. Besides,
having a SoundBlaster does nothing for PBX.
Amancio Hasty Jr
or less if you can read instructions. Or if you are really smart and read the
list of supported hardware to verify that your hardware configuration is
supported the task can be very painless. It is not a big effort
to write a boot floopy which can help you install the free unix systems from
lets say a cdrom.
Yes the freeunix groups are working on it so do expect a big improvement in
the near future.
Again, don't forget we have pearl and tcl to help us write sophisticated
installation tools.
Before anyway starts flaming me I have written installation procedures
for many packages which I have worked on ( I have over 13 years of
experience as a professional software engineer)
Robert Withrow
| Or is this just a subjective, personal opinion? In which case some
| background on how you come to this conclusion would help us all evaluate
| whether we agree with it.
I agree with it from my experience. For example, SVR4 in all its wonderful
variations is chock full of bugs. No vendor has fixed them all. If you
doubt this assertion, please read Eric's bug list.
WRT SCO, I work with a manufacturer who settled on SCO on the basis of
it's maturity, and is frequently finding bugs in various components that
require fixes from SCO. SCO does, however, seem to try harder to fix them
than most vendors. But they are bugs, nonetheless.
WRT DEC's VMS, I used to work fixing bugs in it. 'Nuff said.
WRT the ``free'' oses, some users use them with *no* problems. Does this mean
they have fewer bugs? No. But it does indicate that they are not the
unusable buggy concoctions that some people assert.
With the ``source supplied'' oses, however, if you have the skill you can
fix the bugs. With the ``commercial'' oses the mantra of ``you got technical
support'' would be better stated as ``all you got is technical support''.
And of course, if you don't have any technical skill you are probably lost
either way.
--
Robert Withrow, Tel: +1 617 598 4480, Fax: +1 617 598 4430, Net: wi...@rwwa.COM
R.W. Withrow Associates, 21 Railroad Ave, Swampscott MA 01907-1821 USA
S. Kent Hamilton
> ke...@ksmith.com (Keith Smith) writes:
> >Funny, I missed the soundblaster support in SCO. Is it on sosco? Is
> >it user supported or SCO supported or Card vendor specific?
> Soundblaster? What's that? Oh, you mean one of those little kiddie toy
> things with a joystick port that makes k-k00l "bang-bang" sounds when
> you press the joystick button? No, I haven't seen support for that in
> SCO software... Then again, SCO UNIX isn't a kiddie-game OS. If you
> want that, go run Messy-DOS.
Michael's at it again I see... I find this offensive since I have a
friend that is blind that uses his home computer (and my home Unix
system via dial-up.) quite nicely thanks to his soundcard, and is
even doing a fairly good job of learning to program with it.
If you have nothing nice to say try not saying anything.
--
=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=
Kent Hamilton | Home: ke...@hns.st-louis.mo.US
Sitting in my den | Office: ke...@ops.scscom.COM
St. Peters, MO | Phone: (314) 997-7766 ext 3070
=*=*=*=*=*=*=*= PGP Public key available upon request =*=*=*=*=*=*=*=
If only one could get that wonderful feeling of accomplishment without
having to accomplish anything.
A Wizard of Earth C
>Read the post slightly more carefully. I never stated anywhere that I consider
>Linux or BSD to be worthless. All I did was ask a question, some dumb hick
>fool asks which UNIX systems *do* support Specialix cards? That is a stupid
>question asked by someone obviously showing his ignorance of the marketplace
>in question.
Actually, I am the "dumb hick"; it's true that I am a relative newcomer
to the market ...I have only been programming for 16 years, and have only
been programming UNIX serial software and drivers for 9 years. Oh, and
hacking UNIX kernel code as a hobby for 12 years, and professionally
for 2 1/2 years.
I'm sorry that I didn't recognize your name; I'm located in the US, and
in the past have only dealt with Computone, Digiboard, Arnet, Control
Systems, and several other multiport venders in my 5 years with my previous
employer (whose serial communications product, TERM, beat out UUCP for
number of users according to a UNIX World survey -- if you have a copy,
look for my name in a lot of the files).
>Also *I* have been tasked with finding somebody somewhere who is capable of
>porting our drivers to Linux, during that proces I have also been put in
>touch with somebody who would like to port them to BSD aswell, *THAT* is how
>worthless we see linux and BSD.
That "somebody who would like to port them to BSD aswell" would also be
me; however, since my employer is a major UNIX vender, I had to forward
to the FreeBSD hackers list for someone else to actually hack into the
public sources (I have a non compete agreement).
>It is not that we consider our products so important that we ignore a system
>that doesn't support them. The fact is that *NO* operating system by default
>will support a Specialix Intelligent IO controller, we have to do it ourselves
>with or without the help of the OS vendor.
>
>That fact is true of *ANY* operating system and *ANY* intelligent IO
>controller (to the best of my knowledge).
Yep; which is why I asked my question -- showing my ignorance of the
Specialix portion of the multiport serial board marketplace.
---
"Why do you ask?" "In order to find out."
Stuart Lynne
Brandon S. Allbery <b...@kf8nh.wariat.org> wrote:
>In article <2eqqq0$m...@u.cc.utah.edu>, te...@cs.weber.edu (A Wizard of Earth C) says:
>wrote a driver for it. (SCO cannot reasonably hope to keep up with the
>ludicrous number of different hardware implementations in the PC marketplace.
>Neither can anyone else, Linux and *BSD included.)
What SCO does is to work with major board vendors to ensure that SCO drivers
are available.
In September I bought a new Adaptec 2742 EISA SCSI controller. It came with
a BTLD disk for SCO.
Today I ordered a new 3COM 3C579 EISA Ethernet controller. It doesn't come
with SCO drivers, but if you check ftp.3com.com:/drivers; you'll find
5x9sco.exe which contains a 50 page documentation file for SCO install and
custom installable disk image.
Both of these boards are the manufacturers current state of the art design.
Hot of the press (so to speak). Both are incompatible with previous versions
to the extent of requiring new drivers. Both have SCO drivers available
either with the board in the box at introduction (Adaptec) or via ftp
shortly after introduction (3Com).
Nigel Whitfield
>We no longer support that version; better yet we no longer support
>that product. Or, how about 4 week turn around to just tell you
>that they receive your problem statement and yes they have given it
>a lot of thought and will in the future get back to you.
No. No bells rung there, thanks. If I got support like that, I'd tell
people where to stick it and change to an OS where I could get
sensible responses. How exactly would this be better if I had source
code and an unsupported OS?
Oh yes! My production editor could stop work in press week, post a
query to the net and while they're waiting an indeterminate time for a
response (assuming someone who know's what to do is reading) they can
poke around in the C code to try and fix it themselves. Because, after
all, every business that has a Unix system in the building has a C
programmer, doesn't it?
I can achieve the level of support with which I am confident for the
applications that we are using, and that support can be accessed by
anyone in the office when they need it. They don't need to be
programmers, or Unix gurus. They just need to do their job, with the
computers helping out. That's a commercial reality. It doesn't allow
me to risk thousands of pounds hoping that the goodwill (and I know
it's considerable) of people on the net will come up with a fix.
Unfortunately, this support issue is one of the things that comes up
again and again when people talk about free software. I know lots of
it is great stuff, and wonderful for a lot of things, but there really
are situations where it's not a realistic solution. Contrary to some
of the mail I've received, I do know what I'm talking about when I
decide what's right for my job. More willingness to accept that might
see many people more kindly disposed towards free software.
Nigel.
wpw...@darkwing.austin.ibm.com
>
>In article <2em4ds$n...@vanbc.wimsey.com>, s...@vanbc.wimsey.com (Stuart Lynne) writes:
>
>| Or is this just a subjective, personal opinion? In which case some
>| background on how you come to this conclusion would help us all evaluate
>| whether we agree with it.
>
>they have fewer bugs? No. But it does indicate that they are not the
>unusable buggy concoctions that some people assert.
>
>With the ``source supplied'' oses, however, if you have the skill you can
>fix the bugs. With the ``commercial'' oses the mantra of ``you got technical
>support'' would be better stated as ``all you got is technical support''.
>And of course, if you don't have any technical skill you are probably lost
>either way.
>
And, at least with Linux (I have no experience with other free
386-based OSs), bugs get fixed *fast*. I have seen people report
problems with a certain program on USENET, having the developer reply
to their post with a patch within a day or less. Doesn't happen all
the time, but it seems to happen more than one would expect.
<><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><>
Bill Woodward | wpw...@austin.ibm.com | Crack the Sky fanatic
AIX Graphics Support | 512-838-2834 | Have a day.
My other computer is a 386 running Linux !!!
Pat Spinler
: In article <hastyCI...@netcom.com> ha...@netcom.com (Amancio Hasty Jr) writes:
: >We no longer support that version; better yet we no longer support
: >that product. Or, how about 4 week turn around to just tell you
: >that they receive your problem statement and yes they have given it
: >a lot of thought and will in the future get back to you.
: I can achieve the level of support with which I am confident for the
: applications that we are using, and that support can be accessed by
: anyone in the office when they need it. They don't need to be
: programmers, or Unix gurus. They just need to do their job, with the
: computers helping out. That's a commercial reality. It doesn't allow
: me to risk thousands of pounds hoping that the goodwill (and I know
: it's considerable) of people on the net will come up with a fix.
(Many other comments about commercial vrs frewware support deleted)
During this discussion, no one seems to have mentioned the small but
growing availablity of support contracts from freeware OS's such as
Linux. I am aware of at least three organizations which offer to
provide this on a contract basis. I should note that I use Linux at
home, and have made several proposals to use it at work, with the
addition of such a support contract.
Furthermore, I must agree that many freeware products have fewer bugs
than many commerical products, and those which are there seem to be
fixed in a more timely fashion. As an example, look at any version of
sysVr4 vrs either of the main freeware OS's, Linux or Net/FreeBSD.
This being said, I fear that I must agree with Nigel that the
availablity of commerical support is most often a make or break
decision for a product. Most of the problems I have personally
encountered, or have helped people solve, were not the type that is
easily anemiable to solving by source perusal.
As an example, many problems I encounter are during the stages of
installing or performing esoteric configurations related to hardware.
I can't speak for anyone else, but I am most often not able to read
the source for someone's driver and discern how the hardware is
supposed to behave.
On the other hand, a commercial support organization is likely to have
people on staff intimately familiar with the code in question, or to
have enocountered the question before. Therefore, they are able to
produce the answer before me.
This is not to say that a post on the net will not garner the
attention of the code's authors, but I can't be certain of that. Nor
can I be certain that the author will have the time or resources to
address my questions.
This is perhaps the 'nub' of the issue. If I am paying someone to
support me, I have some assurance that my problem will be addressed,
however competently. If, on the other hand, no one has any obligation
to help me, I may be easily left with a unsoluble problem and no-where
to turn except the unemployment lines. This obligation counts for a
lot of 'cover your ass' in business. To be blunt, it means I have
someone else to blame.
Finally, it may be noted that for certain critical applications
(nuclear power, certain industrial, certain military, etc) it is
possible to obtain support contracts (for many $$$) that garuntee that
your problem will be solved in X period of time, with the provision of
massive damage payments if they are not. Such a contract simply is
not obtainable in the freeware world.
I hope that as the freeware support industry gains acceptance and
maturity, that this situation will change.
L8r,
-- Pat
--
"Practice random kindness and senseless acts of beauty"
--
Pat Spinler "Life ? Life is a crock.
pa...@speedway.net,pspi...@mr.net You're born, you die, and you're
Work: 505/893-4655 Home: 505/294-5923 lonely a lot in between."
Aris Stathakis
>ke...@ksmith.com (Keith Smith) writes:
>>Funny, I missed the soundblaster support in SCO. Is it on sosco? Is
>>it user supported or SCO supported or Card vendor specific?
>Soundblaster? What's that? Oh, you mean one of those little kiddie toy
>things with a joystick port that makes k-k00l "bang-bang" sounds when
>you press the joystick button? No, I haven't seen support for that in
>SCO software... Then again, SCO UNIX isn't a kiddie-game OS. If you
>want that, go run Messy-DOS.
Get TLS022 and TLS023 from ftp.sco.com or via anon-UUCP from sosco. They
include Soundblaster Drivers and software and demos to support the driver.
Aris
--
Aris Stathakis Tel:+27 11 887 4220 Snail Mail:
SCO ACE / Novell CNE Fax:+27 11 786 6647 P.O. Box 78446
M&PD (Pty) Ltd. X25: 06550 11642692 Sandton, 2146
E-Mail: ar...@lasernet.co.za R.S.A.
Wayne Schlitt
err, umm, if support is so critical for you, why don't you *pay* for
it? If you buy your Linux distribution from say Yggdrasil you get a
1-900 number for installation support, and a list of people and places
that will do more extensive support. If you are really hot on
support, you could hire someone to do in house support. It is all a
matter of how critical you think support is.
I know of one business that sells it's $35,000 software package on,
among other systems, netBSD. At that price, the customers expect
things to *work*. They really don't care if the version of YP that
USL got from sun is buggy, and that Dell, or whoever doesn't think
that they can fix it. They can simply charge $1500 per copy of
"supported NetBSD" and then use that money to hire someone full time
to support the system. Believe me, the support they have to do with
NetBSD is less than most other commercial OS's that they use. This
comes out to be a big win all the way around.
Now, you would be crazy to just drop your critical package onto an
untested OS, commercial or free. You would also be crazy to just
write off gcc, Linux, or perl just because it is "free" software. You
need to try these systems and *see* if the work well for you. Then
phase them in as you feel comfortable. The business I referred to
above uses many non-commercial packages simply because no commercial
packages support everything they needed to do. Yacc just couldn't
handle the size of the grammars, so Bison was used instead. Gcc still
supports the 68000, and HP's compilers don't.
-wayne