Linux-Advocacy Digest #577, Volume #25            Thu, 9 Mar 00 20:13:09 EST

Contents:
  Re: JDK1.2.2 performance, Linux -vs- NT (Terry Sikes)
  Re: BSD & Linux (Noah Roberts)
  Re: Motif: Not Invented Here?  (was: The Windows GUI vs. X) (George Richard Russell)
  Re: Thoughts and answers sought for Linux research article ("Francis Van Aeken")
  Re: Microsoft migrates Hotmail to W2K (George Marengo)
  Re: Disproving the lies. (George Marengo)
  Re: BSD & Linux (Alexander Viro)
  Re: Which Linux version is best ? (Steve Mading)
  Re: 11 Days Wasted ON Linux ("ax")
  Re: Why not Darwin AND Linux rather than Darwin OR Linux? (was Re: Darwin or Linux 
(Matt Kennel)
  Re: Salary? (Steve Mading)
  Re: Why not Darwin AND Linux rather than Darwin OR Linux? (was Re: Darwin or Linux 
(Anil T Maliyekke)
  Re: Why not Darwin AND Linux rather than Darwin OR Linux? (was Re: Darwin or Linux 
(John Jensen)
  Re: BSD & Linux (Noah Roberts)

----------------------------------------------------------------------------

From: [EMAIL PROTECTED] (Terry Sikes)
Crossposted-To: comp.lang.java.advocacy
Subject: Re: JDK1.2.2 performance, Linux -vs- NT
Date: 9 Mar 2000 23:12:12 GMT

In article <[EMAIL PROTECTED]>,
Jerry McBride <[EMAIL PROTECTED]> wrote:
>In article <89k8bu$ja$[EMAIL PROTECTED]>,
>[EMAIL PROTECTED] (Terry Sikes) wrote:
>
>-- snip --
>
>>HotSpot is coming in the Linux 1.3 release (due pretty soon I think),
>>and IBM is supposed to ship a 1.3 VM by 2Q.
>
>Any word on whether this will be a free offering or will it be placed on
>Software Choice as a commercial product?

As I understand it, both will be zero cost.

Terry
--
[EMAIL PROTECTED]

------------------------------

From: Noah Roberts <[EMAIL PROTECTED]>
Crossposted-To: 
comp.unix.bsd.386bsd.misc,comp.unix.bsd.freebsd.misc,comp.unix.bsd.misc,comp.unix.bsd.netbsd.misc,comp.unix.bsd.openbsd.misc
Subject: Re: BSD & Linux
Date: Thu, 09 Mar 2000 15:26:57 -0800

I got it working finally last night, took litteraly 10+ hours.  The probalem
was not with either Win98 nor Linux and I can spell out why and the steps it
takes to install the 3 together....many of which should not be necissary in my
opinion.

***DO NOT DO THIS ON A HARD DRIVE WITH DATA YOU WISH TO KEEP, CHANCES ARE HIGH
IT WILL GET LOST OR CORRUPTED!!!*****

You have to use a combination of all 3 fdisks, and you HAVE to do a lot of
writing down of stuff.  See, the NetBSD install system is stupid (not as in a
bad interface, but it gets EVERYTHING wrong).  When it tells you to only say
you have 1024 cylinders, DO NOT believe it...that is mistake #1, causes the
drive to lock up the BIOS if you do not enter in the CORRECT *LBA* geometry for
the drive!  The install says it will get the rest....it lies.

Second major problem, NetBSD fdisk likes to fuck up partition tables.  It
guesses wrong and rewrites them off the end of cylinders.  So, you really need
to install NetBSD first, BUT you need to set up the other partitions beforehand
or nothing will work.  Then after I was finished with everything, NetBSD could
not boot because the superblock was bad.  I did end up with an extended
partition that does not match the geometry at the end, but so far it is working
ok.

Steps to installing win98/BSD/Linux
1. Use *Linux* fdisk to create all partitions.  I was using FAT32 for the win98
partition so this was not a problem.
2. Linux fdisk does not know what a NetBSD partition is, but the type is a9 and
it can create them.
3. Switch to sector view and write EVERYTHING down (remember how I said NetBSD
guesses wrong?  Well this is in preperation for that.)
4. Do the math to find the sizes in sectors for each partition, NetBSD uses
offset and size, not begin/end.

    At this point I had something like the following (forgive the non-tech
layout, I am at school on a win95 box)

    1. 2 gig fat32
    2. 50m Linux native
    3. 3gig unknown (a9/NetBSD)
    4. about a 5 gig extended
    5. various logical drives (NetBSD doesn't see these and you will likely
have to delete the extended anyway as NetBSD is going to fuck it up.)

Now your milage may vary, simply changing the size of each partition a little
bit may make life easier or a lot harder....this was a hack and 2 dozen to get
this working.

Now, boot into NetBSD floppy, run the setup.  When it gets to the partitioning
part...change to sector view.  If you may notice, everything is at least 1
sector off.  Now, you may think it is as simple as changing them all to the
correct sectors you had in linux fdisk....well only if your lucky.  See not
only does NetBSD guess wrong, it doesn't know what IT thinks either.  The first
partition on my drive had to be left alone, even though NetBSD said it ended on
the same exact sector that the next began, and even if your writing in sectors,
there is no way of knowing were NetBSD fdisk is really going to end the
partition...could even be off the end of the disk!

So, do all the changing you think you need, pretend your going to install
NetBSD and wait for it to format all the drives and stuff.  Now reboot into
linux again and check to see if your partitions are still straight....repeat
untill you get it right.

*IMPORTANT* NetBSD can actually change the way the BIOS views your Hard Drive.
It writes some sort of data to the beginning of the drive that gets used by the
BIOS when figuring the geometry in LBA mode.  IF YOU GET IT WRONG when you
first get asked the geometry of your drive in NetBSD you may end up with a
FROZEN BIOS or a 500M HD...usually though you end up with one that is about 1/2
the right size (this is for LARGE disks in excess of 1024 cylinders in LBA
mode)  Enter in the exact information that the BIOS gives in CMOS. (now
something I did NOT try, that you might if you run into trouble is one less
head....like mine said 255, had I entered in 254 results may have been
different, just now thought of it)

If you get a frozen BIOS it is somewhat easy to fix...tell the BIOS to ignore
the drive and boot Linux, which can see it anyway.  Then run the following
command:

dd if=/dev/zero of=/dev/hda bs=512 count=1

Yep, thats right....clear the MBR and start over....fun isn't it.

Ok, I was unable to keep NetBSD from making the extended partition wrong.  If
you delete that partition and try to make a new one with linux fdisk you will
find it can only make one of 0 length (I think the disklabel in NetBSD is
confusing it) so you need to boot into a windows or dos disk and create an
extended partition with DOS fdisk.  This will correct the errors in the
extended partiton.

Now, boot into linux, use fdisk to make the logical drives you want and
install.  Boot into windows and install.  Boot into NetBSD and install (I hope
you wrote all the sector sizes down as seen by NetBSD when it was all working
right! it may have changed now that windows fdisk made a partition)  Yeah, if
you install NetBSD first you can fix the partitions after it fucks them up,
however....I got a bad superblock and unbootable NetBSD system after that so I
doubt it will actually get you anywere.

You may still end up with an extended partition that is 1 sector too big,
however...now that everything else is set up in the logical drives, that should
not cause any problems s far as I can tell.....its only been a few hours and is
not very well tested.

Now, NetBSD can see the linux partition and knows what it is, but is apperently
unable to boot it.  LILO on the other hand can do all 3.  Your partition table
may look something like this:

hda: hda1 hda2 hda3! hda4 <hda5 hda6 hda7 hda8 hda9> <hda10 hda11 hda 12 hda13
hda14>

Simply use liloconfig and tell it to boot another DOS drive (other=) and tell
it to boot off of /dev/hda10 (that is wd0a).

====================================================

It is my decided opinion that this should not be so complicated a process.
Simple fact of the matter is that NetBSD fdisk is somehow non-compatable with
everyone else.  It figures partitions on an odd level and, simply put, destroys
things.  If you were to put it by itself then ok....however I do not really
recomend these *BSD systems for the light of heart.

   I am installing them because I LIKE to play....I am not in a production
environment and do not need my system to be all that stable.  I think you
should be able to get away with NetBSD and windows (though my HD did give up
the goast only days after installing NetBSD...but that is probably a
cooincidence) if you need to play around with it and test it.  I am building
myself a research box, with a BSD and the rest for learning how to program in
different environments....I would like to use VMWare.  NetBSD however IS a
problem system for things like that....don't try it if you haven't a LOT of
time to fuck with it over and over again.


------------------------------

From: [EMAIL PROTECTED] (George Richard Russell)
Subject: Re: Motif: Not Invented Here?  (was: The Windows GUI vs. X)
Reply-To: [EMAIL PROTECTED]
Date: Thu, 09 Mar 2000 23:49:56 GMT

On Thu, 09 Mar 2000 17:15:10 -0500, Donn Miller <[EMAIL PROTECTED]> wrote:
>George Richard Russell wrote:
> 
>> [snip "I wrote it with Motif, and the GUI is large and boring, compared
>> to writing cool new stuff." Fair enough] Otoh, Motif is not the toolkit of
>> choice for new stuff.
>
>I kind of like playing around with Motif, though.  Companies are
>probably still hiring Motif developers, so I've got to be prepared. 
>Which toolkits are commercial users using?  A lot of them still use
>Motif:  VMWare and Corel WP are two examples.  Plus, everytime I see a
>shot of a UNIX workstation on the news (such as CNN), I see a Motif
>app when the camera gives a close-up shot of the screen.

The back of every Linux distro box will have Qt / KDE or GTK / Gnome +
Motif-based Netscape. 

Thoughts - Applixware has moved to GTK from Motif
So has Mozilla.

Corel Office is being ported with Winelib providing the widgets - in
a KDE look and feel. (will be themeable sooner or later)

The Linux Delphi is supposed to target Qt / KDE for Widgets.

Motif stuff is usually ported from where Motif + CDE is ubiquitous.

>I don't know...  I guess it's OK to like more than one toolkit.  I
>will still be using Gtk and Qt, but I'll still be using Motif.  What
>we need is one toolkit that uses "skins".  That way, all our apps will
>look identical, but we can vary the look-n-feel just by changing the
>skin.  I guess that's the same as themes, but it goes futher and
>changes the actual shape of the widgets.  You know, we can make our
>buttons look round instead of rectangular.

Qt 2.0 and greater ;-) Qt 1.x has Motif or Windows look n feel, 2.x
adds a Mac / Platinum one, and KDE extends this to CDE / NeXT / 
Customised / Pixmapped / Gradiented etc.

Future GTK i.e. 1.3.x, 1.4.x may well have more themeing than 1.2.x

George Russell
-- 
One ring to bring them all, and in the darkness bind them.
                                 Lord of the Rings,     J.R.R.Tolkien
Hey you, what do you see? Something beautiful, something free?
                                 The Beautiful People, Marilyn Manson

------------------------------

From: "Francis Van Aeken" <[EMAIL PROTECTED]>
Subject: Re: Thoughts and answers sought for Linux research article
Date: Thu, 9 Mar 2000 20:47:01 -0300

DeAnn Iwan <[EMAIL PROTECTED]> wrote in message news:[EMAIL PROTECTED]...

> suitable for those.  And not for those that run under Linux.  Linux is
> faster and has a more reliable and faster frame rate for things like
> Quake.  But for the games that will only play under windows right now,
> then windows is better.  For everything else....I like linux better.

>From http://cnn.com/2000/TECH/computing/03/09/quake3.linux.idg/index.html :

"Quake III plays normally in full-screen or DGA mode under the XFree86 X Window server.
During a test of game speed on our test machine -- a Penguin Computing Gaming System 
with
a 500 MHz Athlon with 128 MB RAM, 3Dfx Voodoo3 3000, and SoundBlaster PCI 128 --
we got 32.5 frames per second on average running on a 1024 by 768 pixel screen at 16 
color
bits per pixel. This is slower than the speed of a similar Windows machine running 
Quake III,
but it works well enough for playability. "

Who should I believe, LinuxWorld or you?

Francis.




------------------------------

From: George Marengo <[EMAIL PROTECTED]>
Crossposted-To: comp.os.ms-windows.nt.advocacy
Subject: Re: Microsoft migrates Hotmail to W2K
Date: Fri, 10 Mar 2000 00:06:57 GMT

On Thu, 09 Mar 2000 13:37:17 GMT, "Chad Myers" <[EMAIL PROTECTED]>
wrote:
<lots of incorrect information snipped>
>NT 4.0 Was Red book, now (1999) is fully C2 Orange and Red

Please read the definitions of Orange and Red wrt C2 certification. 
In a nutshell, Orange is C2 with no network. Red is Orange with
networking added. 

You don't go from Red to Orange unless your product is getting
 *worse*; i.e. was O.K. to be networked but no longer can do so.


------------------------------

From: George Marengo <[EMAIL PROTECTED]>
Crossposted-To: comp.os.ms-windows.nt.advocacy
Subject: Re: Disproving the lies.
Date: Fri, 10 Mar 2000 00:16:03 GMT

On Thu, 9 Mar 2000 09:58:19 -0500, "Drestin Black"
<[EMAIL PROTECTED]> wrote:

>
>"Nik Simpson" <[EMAIL PROTECTED]> wrote in message
>news:yoOx4.7$py.361@client...
>> Even if you gave away the OS and the database you still don't get even
>>close  to what you claim.
>
>you have the same opinion of rexes crap as I do... what you wrote above is
>what I've been trying to hammer into the linvocates heads and they just
>don't get it. the price of the OS is insignificant except to single PC home
>users (well, less than middle-class income ones at least). 

The price of the OS is important to _any_ business who would have 
to buy not just one, or a few copies of the OS, but several hundred 
or even thousands. I mean, you _are_ talking about legally doing this,
right? That doesn't even include the cost of upgrading the hardware 
so that it can run Win2K.


------------------------------

From: [EMAIL PROTECTED] (Alexander Viro)
Crossposted-To: 
comp.unix.bsd.386bsd.misc,comp.unix.bsd.freebsd.misc,comp.unix.bsd.misc,comp.unix.bsd.netbsd.misc,comp.unix.bsd.openbsd.misc
Subject: Re: BSD & Linux
Date: 9 Mar 2000 19:11:05 -0500

In article <[EMAIL PROTECTED]>,
Noah Roberts  <[EMAIL PROTECTED]> wrote:
>I got it working finally last night, took litteraly 10+ hours.  The probalem
>was not with either Win98 nor Linux and I can spell out why and the steps it
>takes to install the 3 together....many of which should not be necissary in my
>opinion.

[snip litany of horror]

>It is my decided opinion that this should not be so complicated a process.
>Simple fact of the matter is that NetBSD fdisk is somehow non-compatable with
>everyone else.  It figures partitions on an odd level and, simply put, destroys
>things.  If you were to put it by itself then ok....however I do not really
>recomend these *BSD systems for the light of heart.

Hrrrm... Didn't try that with NetBSD, but FreeBSD sysinstall is well-behaving
beast - nothing that would be _that_ nasty. Never tried to add Windows into
the mix (I don't use the thing, and there's nothing to learn from them), but
I doubt that it would change the effect. Triple-boot testbed (kernel stuff
under Linux, testing the behaviour under FreeBSD and Slowlaris in situations
when manpages do not describe the boundary cases well enough) was not a
problem.

Do you, by any chance, have an IBM IDE disk that reports 16383*15*63 geometry?
They got a nice off-by-one - common convention is to report maximal values for
all fields as a sign that disk is too large and another interface should be
used. And maximal number of heads is 16 (15+1, since drive reports heads-1
in 4-bit field). Apparently, IBM decided that they know better and produced
a family of disks that report 16383*_15_*63, instead of common 16383*16*63.
It confused everything that relied on that convention. For Linux the fix
was
+       /*
+        * ... and at least one TLA VBC has POS instead of brain and can't
+        * tell 16 from 15.
+        */
+       if ((id->lba_capacity >= 15481935) && (id->cyls == 0x3fff) &&
+           (id->heads == 15) && (id->sectors == 63)) {
+               id->cyls = lba_sects / (15 * 63); /* correct cyls */
+               return 1;       /* lba_capacity is our only option */
+       }
(2.2.8, IIRC). Check NetBSD fdisk - they may rely on the same thing.

-- 
"You're one of those condescending Unix computer users!"
"Here's a nickel, kid.  Get yourself a better computer" - Dilbert.

------------------------------

From: Steve Mading <[EMAIL PROTECTED]>
Subject: Re: Which Linux version is best ?
Date: 10 Mar 2000 00:25:21 GMT

Denis Barthel <[EMAIL PROTECTED]> wrote:
: darling,
: if i'd have an as/400 or a mainframe, i wouldn't run linux on it.
: linux/unix is the best you can use on a microcomputer-level, but leaving
: this level should mean to leave this os-world ( not talking about
: smp/clustering ).

I wasn't advocating replacing OS/400 with Linux.  I said they can run
both at once.  Admittedly I'm no expert on these computers, but a
write-up by someone who had done this claims that OS/400 isn't any
more "native" on the machine than Linux is - *both* run in virtual
machines provided by the hardware, and (this is the cool part), they
can run at the same time as each other.  Similar to what is done with
VMware (except in hardware), you can run Linux on the same machine
that is running OS/400, at the same time.  Unlike VMware, though,
*both* OSes are actually 'guests' on the host hardware.  The concept
of walled-off virtual machines is actually built-in to the hardware
of AS/400.  When I found that out I was truly impressed.  I had gotten
a bad impression of AS/400s as being boring COBOL crunchers - using
thier fast power merely to compensate for the bloat of the banal
business software that runs on them, but it seems that this reputation
is undeserved.

-- 
-- ------------------------------------------------------------------
 Steven L. Mading  at  BioMagResBank   (BMRB). UW-Madison           
 Programmer/Analyst/(acting SysAdmin)  mailto:[EMAIL PROTECTED] 
 B1108C, Biochem Addition / 433 Babcock Dr / Madison, WI 53706-1544 

------------------------------

From: "ax" <[EMAIL PROTECTED]>
Subject: Re: 11 Days Wasted ON Linux
Date: Fri, 10 Mar 2000 00:32:03 GMT

If someone tells a business owner that he has to learn a lot
in order to use Linux, the business owner will lose interest
on Linux right away.  Business owners are only interested
in getting their daily jobs done taking computers as tools.

If someone tells a business owner that he has to buy
new computers in order to get Linux up and running,
the business owner will give up on Linux since preserving
current technology investment is business owners' high priority.


"gdiv" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> I wish that your experience had been more positive. If you
> were like me and had learned unix before dos/winxx you might
> have the opposite opinion. I remember in 1990, when I first
> started on unix using color sun workstations. Wow, I never
> wanted to log out. I never knew computers could be so much
> fun. I dreamed of the day that I could have my very own
> workstation at home. I didn't want one of those rinky-dink
> pcs. I wanted something fast, powerful, serious, and with a
> sophisticated command line. I was resistive to the m$ empire
> for 9.75 years.
>
> Finally, I got a pc and found myself using windows98 for
> several hours every day. Barf!! That only lasted for a few
> months. Every day I became more dissatisfied with the click
> on this and that and this menu pops up and that wizard pops
> up and all that cartoonish stuff and no *real* windows where
> you can be editing and running *real* programs
> simultaneously, four or more per desktop with four virtual
> desktops at the same time: 16 total (not advised, but
> possible) just like *real* programmers do. I tried real
> programming using the dos prompt and notepad and after about
> six weeks I was ready to kill somebody.
>
> Now I have linux and I am happy. I waited 10 years for
> this!! No one can appreciate the beauty and the power of
> unix/linux until they try to do something that requires its
> special capabilities. It's worth the extra effort. Maybe
> another distribution would have suited you better. Corel
> Linux and Open Linux are easier, from what I understand, and
> have a more m$ish feel to them.
>
> Please don't give up. There's a lot more to it than pain and
> frustration. The rewards are worth it.
>
>
> * Sent from AltaVista http://www.altavista.com Where you can also find
related Web Pages, Images, Audios, Videos, News, and Shopping.  Smart is
Beautiful



------------------------------

From: [EMAIL PROTECTED] (Matt Kennel)
Crossposted-To: comp.sys.next.advocacy,comp.sys.mac.advocacy
Subject: Re: Why not Darwin AND Linux rather than Darwin OR Linux? (was Re: Darwin or 
Linux
Date: 10 Mar 2000 00:40:23 GMT
Reply-To: mbkennel@<REMOVE THE BAD DOMAIN>yahoo.spam-B-gone.com

On 8 Mar 2000 14:06:09 GMT, John Jensen <[EMAIL PROTECTED]> wrote:
:mbkennel@   yahoo.spam-B-gone.com <REMOVETHEBADDOMAIN> writes:
:
:: I have to disagree: because I don't see an opportunity for Apple becoming
:: ``relevant'' in a Linux standard any more than it could be relevant in
:: a Windows standard other than making x86 PC's, in which case it would have
:: preceded Packard Bell into Dogbert's Dumb-pster. 
:
:When I say relevant, I mean that Apple should be able to win when Linux
:wins.  The danger of the antithesis is that any win by Linux would be seen
:to diminish Apple.

Perhaps.  I don't know what they can do about this other than making OSX a
really truly compelling alternative to NT or Solaris and hope that commercial
OSes can hold their own against Linux.  

Notice that even as Linux is `winning', companies involved with it may
not. VAlinux is still tiny versus AAPL. 

:I don't think Apple has to sell dumb clones to be relevant, but they have
:to find a way to say "cool!" a little more often and a little more loudly
:when good things happen in Linux land.

Talk is cheap. 

:: I think their current plan is to work on OS X really hard. 
:
:Absolutely, and this should be their first priority.
:
:: Another muddle is :  Linux is big!  Apple just has to *do something*!  
:
:: What, exactly?   Two things that are clear wins it's already done: QTSS and
:: OpenPlay (network gaming library to compete against Microsoft's).  But
:: I can't figure out any grand strategy. 
:
:Or "Why, exactly?"
:
:Why didn't Apple make a Quicktime player for Linux?
:
:They certainly didn't owe it to anyone, but why did they decide not to
:make friends?

That's a good one, I'd forgotten.  I agree, I want a QT player for
Linux.  But notice this really helps Linux users like me and doesn't do all
that much for Apple yet.  (If Linux takes off more it would, of course). 

I have heard that the problem is not with Apple's code but there are 3rd
parties who own parts of the codecs that are not free.  Still this is
where some encouragement from AAPL could be helpful. 

But like the other things Apple is doing it's a small nice charity
thing here and there and not any Grand Linux Strategy.  {where did
Sgi's grand NT strategy get them?  Like usual, Microsoft got the
goldmine and they got the shaft}

Another thing is WebObjects for Linux, but this eventually may
directly compete with their server hardware offerings. (WO for Solaris
is for bigger hardware than Apple is ever likely to sell) I do believe
that the current feel among web providers is that Linux h/w is now OK
and Apple hardware is not really considered---so WO for Linux would be
generally a win.  Shouldn't be much of a change from their Solaris
port.

:John

-- 
*        Matthew B. Kennel/Institute for Nonlinear Science, UCSD           
*
*      "To chill, or to pop a cap in my dome, whoomp! there it is."
*                 Hamlet, Fresh Prince of Denmark.

------------------------------

From: Steve Mading <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.networking,comp.os.linux.misc
Subject: Re: Salary?
Date: 10 Mar 2000 00:49:05 GMT

In comp.os.linux.advocacy Vilmos Soti <[EMAIL PROTECTED]> wrote:

: Education. If you want to learn, then there are wonderful opportunities
: in the US due to the fact that you can have access to the latest/greatest
: educational tools. But the average US person (I deliberately don't write
: American since it includes Canadians, Mexicans, and a lot of other nations)

True, but there exists no shorthand word for "US-ian", which is why
the word "American" often gets used.  "American" is a lot faster
to say, and it can be argued that since "USA" does have the word
"America" in it, that it kind of fits, in a way.

Its sort of like the way "Asian" is the current politically correct
term for that race of people who live in "the orient".  This term
is being misused since there are other people from the continent of
Asia that are not part of that race - people from Siberia, India,
and the middle-east are also Asian, technically.

-- 
-- ------------------------------------------------------------------
 Steven L. Mading  at  BioMagResBank   (BMRB). UW-Madison           
 Programmer/Analyst/(acting SysAdmin)  mailto:[EMAIL PROTECTED] 
 B1108C, Biochem Addition / 433 Babcock Dr / Madison, WI 53706-1544 

------------------------------

From: Anil T Maliyekke <[EMAIL PROTECTED]>
Crossposted-To: comp.sys.next.advocacy,comp.sys.mac.advocacy
Subject: Re: Why not Darwin AND Linux rather than Darwin OR Linux? (was Re: Darwin or 
Linux
Date: 10 Mar 2000 00:17:59 GMT

In comp.sys.next.advocacy Charles W. Swiger <[EMAIL PROTECTED]> wrote:
> Agreed.  But for Apple to live with clones, they'd have to ensure that both
> their hardware and their software aspects are competitive, and they'd have to
> cut their margins, particularly on generic items like peripherals and memory
> and so forth.  I don't see that happening.

> -Chuck

Wide open cloning no doubt would be a bad idea if Apple intended to remain
a hardware maker.  Choosing one or two licensees who had an installed base
to sell to would work better.  Especially if they have other goodies they
would be willing to bring over with them.

Anil


------------------------------

From: John Jensen <[EMAIL PROTECTED]>
Crossposted-To: comp.sys.next.advocacy,comp.sys.mac.advocacy
Subject: Re: Why not Darwin AND Linux rather than Darwin OR Linux? (was Re: Darwin or 
Linux
Date: 10 Mar 2000 01:01:33 GMT

mbkennel@   yahoo.spam-B-gone.com <REMOVETHEBADDOMAIN> writes:
: On 8 Mar 2000 14:06:09 GMT, John Jensen <[EMAIL PROTECTED]> wrote:

: :I don't think Apple has to sell dumb clones to be relevant, but they have
: :to find a way to say "cool!" a little more often and a little more loudly
: :when good things happen in Linux land.

: Talk is cheap. 

Don't take me too literally ... but I think expressing their own
involvement is one way to declare their relevance.

John

------------------------------

From: Noah Roberts <[EMAIL PROTECTED]>
Crossposted-To: 
comp.unix.bsd.386bsd.misc,comp.unix.bsd.freebsd.misc,comp.unix.bsd.misc,comp.unix.bsd.netbsd.misc,comp.unix.bsd.openbsd.misc
Subject: Re: BSD & Linux
Date: Thu, 09 Mar 2000 16:50:54 -0800

>
> Hrrrm... Didn't try that with NetBSD, but FreeBSD sysinstall is well-behaving
> beast - nothing that would be _that_ nasty. Never tried to add Windows into
> the mix (I don't use the thing, and there's nothing to learn from them), but
> I doubt that it would change the effect. Triple-boot testbed (kernel stuff
> under Linux, testing the behaviour under FreeBSD and Slowlaris in situations
> when manpages do not describe the boundary cases well enough) was not a
> problem.

Well, to tell the truth, win98 is there for my games, though I will want to be able
to write SOME applications that will work in win95/98 that will be ported versions
of the other stuff....

>
>
> Do you, by any chance, have an IBM IDE disk that reports 16383*15*63 geometry?
> They got a nice off-by-one - common convention is to report maximal values for
> all fields as a sign that disk is too large and another interface should be
> used. And maximal number of heads is 16 (15+1, since drive reports heads-1
> in 4-bit field). Apparently, IBM decided that they know better and produced
> a family of disks that report 16383*_15_*63, instead of common 16383*16*63.
> It confused everything that relied on that convention. For Linux the fix
> was

Western Digital Caviar.  10 gig in decimal gigs.  The LBA geometry for it is 1247 C,
255 H, and 63 S.  I did the math, and the LBA is a little short as it has to round a
bit...but I only lose a couple of K.  I didn't try actual geometry with this disk
because the last drive I tried that on got all screwy and actually locked up the
BIOS, putting in LBA seems to at worst result in an unbootable drive that confuses
the BIOS enough that it won't boot off of a floppy....I can still get into setup
though.  Yeah, no it reports 16 heads under normal mode.

I am sure there is a better way of doing it then I did....or maybe I ran into unusal
troubles....but yes, it was a nightmare.  Unfortunately I did try to do this on a
drive that had info on it I wanted (win98 all set up and 'working') and I lost it.
NetBSD does some wierd things to the system that had to be worked around a great
deal.  What really surprised me most of all I think, is when I realised that the
BIOS is actually depending on information NetBSD overwrote, until last night I
didn't think the BIOS talked to the drive that much during the POST sequence to make
it matter what any system did to it.

I may give FreeBSD another shot someday, right now I am too scared to touch what I
have working I think.



------------------------------


** FOR YOUR REFERENCE **

The service address, to which questions about the list itself and requests
to be added to or deleted from it should be directed, is:

    Internet: [EMAIL PROTECTED]

You can send mail to the entire list (and comp.os.linux.advocacy) via:

    Internet: [EMAIL PROTECTED]

Linux may be obtained via one of these FTP sites:
    ftp.funet.fi                                pub/Linux
    tsx-11.mit.edu                              pub/linux
    sunsite.unc.edu                             pub/Linux

End of Linux-Advocacy Digest
******************************

Reply via email to