Linux-Advocacy Digest #950, Volume #30 Sun, 17 Dec 00 22:13:02 EST
Contents:
Re: Sun Microsystems and the end of Open Source ("Chad C. Mulligan")
Re: Sun Microsystems and the end of Open Source ("Chad C. Mulligan")
Re: Sun Microsystems and the end of Open Source ("Chad C. Mulligan")
Re: i LOVE this- the auther is a genius ("Chad C. Mulligan")
Re: Conclusion ("Chad C. Mulligan")
Re: Linux is awful (Johannes Tanzler)
Re: Conclusion ("Chad C. Mulligan")
Re: Conclusion ("Chad C. Mulligan")
Re: Conclusion ("Erik Funkenbusch")
Re: Linux is awful ("Erik Funkenbusch")
Re: Is Windows an operating system like Linux? ("Erik Funkenbusch")
Re: Predictions (featuring Drestin Black) ([EMAIL PROTECTED])
Re: Is Windows an operating system like Linux? (mlw)
----------------------------------------------------------------------------
From: "Chad C. Mulligan" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.ms-windows.nt.advocacy
Subject: Re: Sun Microsystems and the end of Open Source
Date: Mon, 18 Dec 2000 02:10:42 GMT
"Pan" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
>
>
> "Chad C. Mulligan" wrote:
> >
> > "Pan" <[EMAIL PROTECTED]> wrote in message
> > news:[EMAIL PROTECTED]...
> > >
> > >
> > > "Chad C. Mulligan" wrote:
> > > >
> > > > Au contraire, Gary, it is you who is confused. No movement for two
> > years on
> > > > Mozilla,
> > >
> > > I guess that beta version that I recently installed on my computer
with
> > > Komodo is an illusion?
> > >
> >
> > What stage was that? Alpha? Beta? Release?
> >
> > > > the same for the 2.4 Kernel,
> > >
> > > ...2.401 snapshot is installed on one of my pc's along with Mandrake
7.2
> > > and running just fine, thanks.
> > >
> >
> > What stage was that? Alpha? Beta? Release?
>
> It's a development snapshot. Call it whatever you want. It works fine
> for me.
>
> > Still waiting to hear why all the marbles for OSS are a year or more
behind
> > schedule. Can't run a business that way, maybe that's why RH is failing.
>
> I'm not sure who created the schedule you are referring to. The
> timetable doesn't make much sense to me. Certainly it has come no
> slower than I anticipated it. Out of curiousity...
>
To be honest I has come no slower than I anticipated (predicted?) either.
But then with you guys it's always gonna be supported in the next version.
Well with the kernel the next version was promised last year by the almighty
Linus himself. Maybe it's late because Charles Shultz died and there's no
one left to write Linus' code for him.
> Do you find it ironic that you spend many hours fudding linux and yet
> manage to complain when people attribute malicious intent to the
> corporate culture you seem so hell bent on promoting?
>
The Corporations and the Penguinista's are their own worst enemies. They,
together are guaranteeing that OSS will never be more than a passing hobby
in most cases. Businesses don't, nor will they ever, risk their bottom line
on an OS that they cannot guarantee support for. MS's support may be
convoluted or even, in some cases, expensive but it is there and available.
I cannot guarantee that the writer of a given module in Linux will be
available or even willing to assist. My experience in this has indicated
that they won't help.
> --
> [EMAIL PROTECTED]
> http://salvador.venice.ca.us
------------------------------
From: "Chad C. Mulligan" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.ms-windows.nt.advocacy
Subject: Re: Sun Microsystems and the end of Open Source
Date: Mon, 18 Dec 2000 02:12:00 GMT
"Gary Hallock" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> "Chad C. Mulligan" wrote:
>
> >
> >
> > Not a big enough market.
>
> In other words, no one in their right mind would want to run NT on S/390.
> But there are lots of people who want to run Linux on S/390. I wonder
why?
>
Define "lots" and list a few examples outside of academia or IBM's research
divisions.
> Gary
>
------------------------------
From: "Chad C. Mulligan" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.ms-windows.nt.advocacy
Subject: Re: Sun Microsystems and the end of Open Source
Date: Mon, 18 Dec 2000 02:12:59 GMT
"Pan" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> "Chad C. Mulligan" wrote:
>
> > Alpha, Beta and pre Alpha all equal vapourware.
>
> Vapourware is when a product does not exist. With open office, binaries
> and source are being distributed through magazines via cd as well as via
> ftp along with the source. Everyone who is interested *knows* what is
> its status.
>
> Vapourware isn't even a possibility in that environment.
>
> Vapourware is when you've got nothing.
>
> Vapourware is a corporate product.
>
No, vapourware is sucking people into using beta code and never releasing
the product. There is no release so there is no product. Beta == vapour.
> --
> [EMAIL PROTECTED]
> http://salvador.venice.ca.us
------------------------------
From: "Chad C. Mulligan" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.ms-windows.nt.advocacy,comp.os.ms-windows.advocacy
Subject: Re: i LOVE this- the auther is a genius
Date: Mon, 18 Dec 2000 02:13:51 GMT
"Peter Hayes" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> On Sun, 17 Dec 2000 20:43:08 GMT, "Chad C. Mulligan"
> <[EMAIL PROTECTED]> wrote:
>
> >
> > "mlw" <[EMAIL PROTECTED]> wrote in message
> > news:[EMAIL PROTECTED]...
> > > Chad Myers wrote:
> > > >
> > > > "glitch" <[EMAIL PROTECTED]> wrote in message
> > > > news:[EMAIL PROTECTED]...
> > > > > based on my typing skills for the subject im not the 'auther' of
the
> > > > > article, lol
> > > > >
> > > > > glitch wrote:
> > > > > >
> > > > > > http://www.msnbc.com/news/503816.asp
> > > >
> > > > <sigh> Gates doesn't have a conscience, huh?
> > > >
> > > > Never mind the billions he's given away in the name
> > > > fo world heath standards and lessening the suffering
> > > > of millions around the world.
> > > >
> > > > Let's also not mention that he has contributed hundreds
> > > > of millions to local childrens and children-benefit organizations
> > > > around the country to support inner-city and underprivileged
children.
> > > >
> > > > These authors are a bunch of windbags who are nothing more
> > > > than jealous.
> > >
> > > It is very easy to be charitable with ill gotten gains.
> > >
> > > Think about all the companies he has ruined with dirty tricks. Think
> > > about the workers and investors that have lost billions because of his
> > > anti-competitive behavior. Think about all the lost productivity from
> > > his crappy software.
> >
> > Name one incident where the respondent didn't leave the table rich.
>
> Spyglass
>
Doesn't count, the software was public domain. NCSA is public funded
therefore the products are public domain.
> Peter
>
------------------------------
From: "Chad C. Mulligan" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.ms-windows.advocacy,comp.os.ms-windows.nt.advocacy
Subject: Re: Conclusion
Date: Mon, 18 Dec 2000 02:14:24 GMT
"Joseph T. Adams" <[EMAIL PROTECTED]> wrote in message
news:91jie8$2l$[EMAIL PROTECTED]...
> In comp.os.linux.advocacy Erik Funkenbusch <[EMAIL PROTECTED]> wrote:
> : "Adam Ruth" <[EMAIL PROTECTED]> wrote in message
> : news:91buii$10qg$[EMAIL PROTECTED]...
> :> Let me make it more clear:
> :>
> :> The most important factor IN DETERMINING STABILITY is my own
experience.
> : In
> :> other words: Regardless of what any tests, benchmarks, etc say, my own
> :> experience with the systems matters more. Some bonehead on a newsgroup
> : can
> :> tell me that they have been able to keep WinNT running for 3 years
> : straight.
> :> Well, my own experience doesn't bear that out, and that sure matters
much
> :> more than what may or may not have happened in some IT shop somewhere.
> :>
> :> I trust my experience, everything else serves only as a guide.
>
> : What exactly *IS* your experience in attempting to maintain a high
> : availability NT system?
>
>
> Why would anyone bother when better alternatives are available?
>
So your answer would be NONE.
>
> Joe
------------------------------
From: [EMAIL PROTECTED] (Johannes Tanzler)
Crossposted-To:
comp.os.linux,comp.os.linux.x,comp.os.ms-windows,comp.os.ms-windows.advocacy,alt.os.linux,alt.os.linux.mandrake
Subject: Re: Linux is awful
Date: Mon, 18 Dec 2000 03:01:14 +0100
Reply-To: [EMAIL PROTECTED]
On Sun, 17 Dec 2000 22:09:52 +0200, Ayende Rahien wrote:
> > I second this. I couldn't remember what to look for what I eventually
> > got things working, although I remember seeing some interesting looking
> > stuff during the install!
>
> Debian needs it desperately.
> So many packages, so little time...
ACK. I've spent a whole afternoon in dselect when I installed Debian
on my machine.
--
"...you might as well skip the Xmas celebration completely, and
instead sit in front of your linux computer playing with the
all-new-and-improved linux kernel version." (By Linus Torvalds)
------------------------------
From: "Chad C. Mulligan" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.ms-windows.advocacy,comp.os.ms-windows.nt.advocacy
Subject: Re: Conclusion
Date: Mon, 18 Dec 2000 02:19:13 GMT
"Adam Ruth" <[EMAIL PROTECTED]> wrote in message
news:91jkjb$l5k$[EMAIL PROTECTED]...
> Good point, I must acquiesce to your statement. Let's then add the 4th to
> my list:
>
> 4) Companies that use Windows (not windows itself) attracts poor quality
> sys admins, because they can't afford anything else. Companies that use
> Unix attract higher quality sys admins because they can afford them.
>
This doesn't ring true. IAC, I get paid as well as an NT admin as I did as
a UNIX admin. The overall costs of NT are less than commercial UNIX and I
have a testimonial from a major company that they won't use Linux because
the costs of administration have proven too high, they use NT.
> There is deffinitely a truth behind that. The developer of C++, Bjarne
> Stroustrup, commented that when C++ was first released he was pleased at
the
> quality of the developers discussing it on the newsgroups. But as time
went
> on, and C++ became more popular, the quality of the discussions dropped.
> His conclusion was now that C++ had reached a certain level of popularity,
> the average quality of the developing community dropped as a whole. It
was
> a matter of scale.
>
> This too would seem to be true for Windows. Its mainstream use means that
> the quality of people using and administering it is on average lower than
> for those operating systems with smaller user bases. That would seem to
> make logical sense.
>
> How then, can we apply this to the numbers from Netcraft? That's a
question
> that I don't think could be authoritatively answered, with the data as
> presently presented. What we need is a sample of the best performing sites
> from each operating system, at or above the standard deviation from mean.
> That would filter out these many companies that truly don't know what they
> are doing, and give us a good apples to apples comparison.
>
Since the collection method for the Netcraft numbers is severly flawed they
should not be used to cite any significance.
> Since that's probably not going to happen, we're going to have to wait
until
> W2K has been around long enough to give it a chance to be in the list of
top
> performing sites. If it makes it, great, if not, well then we're back to
> square one.
>
> So allow me to ammend my conclusions:
>
> Netcraft's data are insufficient to be authoritative on the stability of
> Windows. We do have W2K data, but not over a very long period, time will
> tell. The NT data are intermixed both with Windows 9x data, limited to
just
> under 50 days, and are heavily influenced by numerous sites that are
> operated by low quality sys admins.
>
Yup, don't forget lazy sys admins, and please ammend low quality to read
under trained.
> Knowing this, we can at least get quality numbers for Unix, in most if its
> various flavors. The interested oberserver would then need to, at this
> point, collate those data with their own experiences and of those they can
> trust. We have, unfortunately, no other data, so each prudent reader must
> fill in the gaps themselves.
>
> In my own case, Unix appears much more stable than Windows. The local NT
> user's group cannot provide any evidence to back up any claim to the
> contrary. There are anectdotal data here in the newgroup, but I give them
> no weight at all. In the end, I can either go with an OS that I know will
> be stable (Unix) or I can go with an OS that may be stable (Windows). I'm
> taking the sure bet.
>
> A quick survey:
>
> What are everyone else's conclusions? I'm interested to read them.
>
I've already posted mine. I find most UNIX implementations stable when
properly administered and the same results with NT. The difference is that
NT has a few capabilities that most UNIX don't.
> Adam Ruth
>
> "Chad C. Mulligan" <[EMAIL PROTECTED]> wrote in message
> news:jq9%5.35872$[EMAIL PROTECTED]...
> >
> > "Adam Ruth" <[EMAIL PROTECTED]> wrote in message
> > news:91g8t7$16ns$[EMAIL PROTECTED]...
> > > I never claimed they where "scientific" (whatever that means, there
> aren't
> > > very many words abused more than this one). Of course there are
better
> > > surveys, and better methods, this is not in doubt (when Nielsen
conducts
> > an
> > > uptime survey, then perhaps that should satisfy more people). But we
> only
> > > have these two sets of numbers, and they are not totally random.
These
> > > numbers were not pulled out of someone's hat. To completely discount
> them
> > > is foolish, they do have some weight. Not perfect, I agree, but they
> are
> > > instructive. What I did claim is that they identified trends.
> > >
> > > So let's look at the trends:
> > >
> > > >> Unix trends toward being more stable than Windows.
> > > > No, Unix admins tend to know their system better than Windows
> > > administrators.
> > > > To even get basic things done in Unix, you have to have a much
higher
> > > level
> > > > of computer knowledge to get basic things done in Windows. Setting
up
> a
> > > > DHCP server would be a basic example.
> > >
> > > This statement clearly implies that one reason Netcraft numbers for
> > Windows
> > > are lower is because Windows sys admins, as a whole, are less
> experienced
> > > than Unix sys admins. But at the same time, Unix requires more work
and
> > > knowledge than an equivalent Unix system. Okay, let's go with that
> > > assumption. It can mean one of three things:
> > >
> > > 1) Windows attracts poor quality sys admins.
> > > 2) Unix forces a sys admin to be high quality, and makes Windows sys
> > > admin's lazy.
> > > 3) There are many more Windows boxes out there, and it's harder to
get
> > good
> > > Windows people because of shortages of talent.
> > >
> >
> > Actually the problem is that a majority of the NT SysAdmins out there
are
> > like you, doing it as a sideline to their regular job. In my experience
> > good programmers do not make good sysadmins. The mindset is different.
> > Conversly good sysadmins aren't necessarily good programmers either. On
> the
> > whole Windows does rather well when you consider that, in smaller
> companies
> > (And that's the primary user base at this time) professional sysadmins
are
> > rare.
> >
> > The administrator is often a technician from the company they bought the
> > system from and on tight time pressure to not spend the proper amount of
> > time on any one system, including unfortunately a customer's server.
> This,
> > and poor accountability, leads to improper practices, short cuts to
> solving
> > problems like everyone being an administrator, and these guys when
called
> > for phone support are likely to tell the user to reboot in the chance
that
> > it'll fix the specific problem instead of talking them through some
> > complicated troubleshooting. The new remote administration tools should
> > aleviate some of these difficulties.
> >
> > The essence remains a properly administered NT system is as stable as
any
> > UNIX. At this time getting the proper administration skills to the
system
> > when they are needed is the problem.
> >
> >
> > > There may be more, I'm sure I'll think of one more 10 seconds after I
> post
> > > this message.
> > >
> > > Anyway, those are three trends that can be identified from Netcraft
> > numbers,
> > > based on the assumption that I inferred from your statement.
> > >
> > > Those trends tell me one thing: Don't use Windows. If I will have a
> > harder
> > > time hiring a quality Windows sys admin than a Unix sys admin,
shouldn't
> > > that be a consideration of mine?
> > >
> > > Now, of course, I expect you to have a different interpretation, and
> I'll
> > be
> > > glad to listen to it.
> > >
> > > Adam Ruth
> > >
> > >
> >
> >
>
>
------------------------------
From: "Chad C. Mulligan" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.ms-windows.advocacy,comp.os.ms-windows.nt.advocacy
Subject: Re: Conclusion
Date: Mon, 18 Dec 2000 02:20:12 GMT
"Ayende Rahien" <[EMAIL PROTECTED]> wrote in message
news:91jmo6$kaa$[EMAIL PROTECTED]...
>
> "Adam Ruth" <[EMAIL PROTECTED]> wrote in message
> news:91jkjb$l5k$[EMAIL PROTECTED]...
> > Good point, I must acquiesce to your statement. Let's then add the 4th
to
> > my list:
> >
> > 4) Companies that use Windows (not windows itself) attracts poor
quality
> > sys admins, because they can't afford anything else. Companies that use
> > Unix attract higher quality sys admins because they can afford them.
>
> 5) Companies that use Windows don't bother to hire full-time sys admins.
> Which leads to users playing at being administrators without the knowledge
> they need to. You *can't* play with linux as root without having the
proper
> knowledge, the system is too complex to let you do this. And if you aquire
> some small knowledge you'll kill the system totally so fast that it
wouldn't
> have time to be unstable. On general, I would say that Windows systems can
> be more stable under ignorant users than a *nix, and as stable as a *nix
> under compotent administrators.
>
> Anyone else encountered users doing this rm /tmp ?
>
or rm -r .* (note in older System V this is recursive both down and up the
directory tree.)
>
>
------------------------------
From: "Erik Funkenbusch" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.ms-windows.advocacy,comp.os.ms-windows.nt.advocacy
Subject: Re: Conclusion
Date: Sun, 17 Dec 2000 20:25:22 -0600
"Adam Ruth" <[EMAIL PROTECTED]> wrote in message
news:91jkjb$l5k$[EMAIL PROTECTED]...
> In my own case, Unix appears much more stable than Windows. The local NT
> user's group cannot provide any evidence to back up any claim to the
> contrary. There are anectdotal data here in the newgroup, but I give them
> no weight at all. In the end, I can either go with an OS that I know will
> be stable (Unix) or I can go with an OS that may be stable (Windows). I'm
> taking the sure bet.
If it's such a sure bet, why are there so many Unix systems with worse
uptimes than NT?
BTW, you will note that there are no Linux machines in the top 50 uptimes.
------------------------------
From: "Erik Funkenbusch" <[EMAIL PROTECTED]>
Crossposted-To:
comp.os.linux,comp.os.linux.x,comp.os.ms-windows,comp.os.ms-windows.advocacy,alt.os.linux,alt.os.linux.mandrake
Subject: Re: Linux is awful
Date: Sun, 17 Dec 2000 20:31:37 -0600
"Adam Schuetze" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]
> You know, the one thing I think is missing from this
> installation procedure (and probably others, I haven't used many
> others), is the capability to save a record of the list of
> packages you selected. There is SO much stuff no that cd.
> Would be nice to be able to keep a record on floppy or
> something. That way, if you want to install again later (or on
> multiple machines) you can use this record from floppy to
> simplify the installation across multiple machines.
Actually, I think Linux would do much better if they just created a simple
base install that got the system up and working with base services, then
allowed you to have a nice utility to install supplementary packages later.
Sure you can do this by hand, and some distro's even give you some crude
tools, but it's not the way that they expect you to install it. Everytime
i've seen a distro that offered a utility to install packages after the
install, it was basically a hacked in version of the original install,
rather than an extended utility which gives more information.
------------------------------
From: "Erik Funkenbusch" <[EMAIL PROTECTED]>
Subject: Re: Is Windows an operating system like Linux?
Date: Sun, 17 Dec 2000 20:45:54 -0600
"mlw" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> Erik Funkenbusch wrote:
> >
> > "mlw" <[EMAIL PROTECTED]> wrote in message
> > news:[EMAIL PROTECTED]...
> > > I don't consider Mr. Schullman an OS expert. I consider him a DOS,
> > > Windows, x86 expert. His opinions outside the pure realm of Windows
are
> > > not substancially athoritative.
> > >
> > > The running DOS in a virtual machine does not an OS make. QEMM,
Pharlap,
> > > and all the other DOS extenders did that. They all have to do that
> > > because interrupts and DMA have to be virtualized. What makes Windows
> > > any different than DesqView?
> > >
> > > DesqView, a DOS extender, provided APIs for applications, multitasking
> > > virtual DOS sessions, and the original DOS which was booted is run in
> > > its own VM.
> >
> > DesqView doesn't provide a file system, it's own driver models, GUI,
device
> > independant graphics subsystems, printing subsystems, memory mapped
files,
> > virtual memory, or any other thousands of other API's (such as
Telephony,
> > 3D, etc..).
>
> These are simply APIs that have been developed over years. Arguments
> about magnitude do not negate the basic assumptions.
>
> Any and all of these APIs could have been implemented in DesqView.
> Windows requires DOS, pure and simple, thus it is not an operating
> system.
They could have been implemented in DesqView, but weren't. Thus, DesqView
is not an OS.
Tell me, would Linux be an OS if it only provided the same services that
DesqView did (from a general perspective, not a DOS perspective). You're
only argument is that if the OS depends on DOS, it's not an OS. That's not
an argument.
> > If you're looking at an OS as a kernel only, then what do you consider
> > mkLinux? Which runs linux in a subsystem?
>
> Yes, mklinux is not the OS in this case.
Alright, the Linux running on a S/390 isn't an OS either. Neither is MacOS
X, nor is the GNU HURD. Your definition of an OS is outdated and certainly
not supported by many experts in the field.
> > > BTW Windows is still based on DPMI, "DOS Protected Mode Interface."
> >
> > Windows provides DPMI to DOS apps, but then so did OS/2. Are you going
to
> > suggest that OS/2 is also DOS based (Strangely enough, OS/2 used a
slight
> > variation of the DOS driver model, which was 16 bit. It was designed to
> > allow DOS drivers to be recompiled with minimal effort).
>
> I did driver development in OS/2 1.x and some app work in 2.x. The
> reason, in 1.x, why drivers were so horrible was because of the DOS box
> and the 286. It was very inefficient to go from protected mode to real
> mode, so interrupts could occur in either mode. A driver writer's
> nightmare.
Indeed, but they didn't have to model the drivers after the DOS driver
model. That was because OS/2 was originally supposed to be DOS 5.
> Under Windows, DPMI is the methodology on which the shell is based. On
> OS/2 DPMI is an API provided by the OS for DOS applications. Do you
> understand the difference?
It's the same with Windows. Only DOS apps can make use of DPMI. Windows
implements it's memory management in a way that is compatible with DPMI, but
to suggest that windows uses DPMI itself is silly, and stupid.
------------------------------
From: [EMAIL PROTECTED]
Crossposted-To: comp.os.ms-windows.nt.advocacy,alt.linux.sux
Subject: Re: Predictions (featuring Drestin Black)
Reply-To: [EMAIL PROTECTED]
Date: Mon, 18 Dec 2000 02:54:29 GMT
On Sun, 17 Dec 2000 19:22:34 GMT, [EMAIL PROTECTED] (The
Ghost In The Machine) wrote:
>And this means Linux is in major trouble?
>
>Spot The Flaw.
The flaw is that major corporations want a total service solution, not
a support structure that is internet based.
Also if Redhat a major player is having problems and Corel are having
problems what does that say about Open Source and the concept of Linux
in general?
It says, there is little money to be made.
Flatfish
Why do they call it a flatfish?
Remove the ++++ to reply.
------------------------------
From: mlw <[EMAIL PROTECTED]>
Subject: Re: Is Windows an operating system like Linux?
Date: Sun, 17 Dec 2000 22:01:30 -0500
Erik Funkenbusch wrote:
>
> "mlw" <[EMAIL PROTECTED]> wrote in message
> news:[EMAIL PROTECTED]...
> > Erik Funkenbusch wrote:
> > >
> > > "mlw" <[EMAIL PROTECTED]> wrote in message
> > > news:[EMAIL PROTECTED]...
> > > > I don't consider Mr. Schullman an OS expert. I consider him a DOS,
> > > > Windows, x86 expert. His opinions outside the pure realm of Windows
> are
> > > > not substancially athoritative.
> > > >
> > > > The running DOS in a virtual machine does not an OS make. QEMM,
> Pharlap,
> > > > and all the other DOS extenders did that. They all have to do that
> > > > because interrupts and DMA have to be virtualized. What makes Windows
> > > > any different than DesqView?
> > > >
> > > > DesqView, a DOS extender, provided APIs for applications, multitasking
> > > > virtual DOS sessions, and the original DOS which was booted is run in
> > > > its own VM.
> > >
> > > DesqView doesn't provide a file system, it's own driver models, GUI,
> device
> > > independant graphics subsystems, printing subsystems, memory mapped
> files,
> > > virtual memory, or any other thousands of other API's (such as
> Telephony,
> > > 3D, etc..).
> >
> > These are simply APIs that have been developed over years. Arguments
> > about magnitude do not negate the basic assumptions.
> >
> > Any and all of these APIs could have been implemented in DesqView.
> > Windows requires DOS, pure and simple, thus it is not an operating
> > system.
>
> They could have been implemented in DesqView, but weren't. Thus, DesqView
> is not an OS.
The quantity of API does not define an OS.
>
> Tell me, would Linux be an OS if it only provided the same services that
> DesqView did (from a general perspective, not a DOS perspective). You're
> only argument is that if the OS depends on DOS, it's not an OS. That's not
> an argument.
I disagree completely. Even if Linux did not support the things it does,
it is an OS because it has the ability to boot on its own and function
without the aid of software which is not native to its kernel.
Windows is a dos extender, a BIG dos extender, but it is a DOS extender.
>
> > > If you're looking at an OS as a kernel only, then what do you consider
> > > mkLinux? Which runs linux in a subsystem?
> >
> > Yes, mklinux is not the OS in this case.
>
> Alright, the Linux running on a S/390 isn't an OS either. Neither is MacOS
> X, nor is the GNU HURD. Your definition of an OS is outdated and certainly
> not supported by many experts in the field.
Which experts would that be? In which field? Windows guys need to have
validation so they say Windows is an OS, but most serious experts don't
think it is.
>
> > > > BTW Windows is still based on DPMI, "DOS Protected Mode Interface."
> > >
> > > Windows provides DPMI to DOS apps, but then so did OS/2. Are you going
> to
> > > suggest that OS/2 is also DOS based (Strangely enough, OS/2 used a
> slight
> > > variation of the DOS driver model, which was 16 bit. It was designed to
> > > allow DOS drivers to be recompiled with minimal effort).
> >
> > I did driver development in OS/2 1.x and some app work in 2.x. The
> > reason, in 1.x, why drivers were so horrible was because of the DOS box
> > and the 286. It was very inefficient to go from protected mode to real
> > mode, so interrupts could occur in either mode. A driver writer's
> > nightmare.
>
> Indeed, but they didn't have to model the drivers after the DOS driver
> model. That was because OS/2 was originally supposed to be DOS 5.
Point of fact, OS/2 was originally intended to be DOS 3.x, but it wasn't
ready, then 4.x, but it wasn't ready, and so on. The 286 DOS box was a
big, huge, problem. IBM insisted that OS/2 1.x run on a 286 because
they, as yet, did not have a 386 based system.
>
> > Under Windows, DPMI is the methodology on which the shell is based. On
> > OS/2 DPMI is an API provided by the OS for DOS applications. Do you
> > understand the difference?
>
> It's the same with Windows. Only DOS apps can make use of DPMI. Windows
> implements it's memory management in a way that is compatible with DPMI, but
> to suggest that windows uses DPMI itself is silly, and stupid.
Take a look low down dude. The Windows executive is a DPMI environment
and Windows run in a virtual DOS machine within it. BTW: Windows runs in
the same virtual machine as the actual DOS OS because each Windows
program has to have a DOS PSP.
--
http://www.mohawksoft.com
------------------------------
** 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 by posting to comp.os.linux.advocacy.
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
******************************
