Linux-Advocacy Digest #608, Volume #28 Thu, 24 Aug 00 03:13:05 EDT
Contents:
Re: Windows stability: Alternate shells? ("Erik Funkenbusch")
Re: Malloy digest, volume 2576+8695i ("Slava Pestov")
Re: Would a M$ Voluntary Split Save It? ("Erik Funkenbusch")
----------------------------------------------------------------------------
From: "Erik Funkenbusch" <[EMAIL PROTECTED]>
Subject: Re: Windows stability: Alternate shells?
Date: Thu, 24 Aug 2000 01:45:50 -0500
"R.E.Ballard ( Rex Ballard )" <[EMAIL PROTECTED]> wrote in message
news:8o0cue$6cd$[EMAIL PROTECTED]...
> > Yes, the first OS that they developed.
> > Still it was a liscensed Sys V port
> > if I recall correctly.
>
> Close - Sys V came out in 1985. Xenix was a variant of Version 6
> UNIX which was distributed among the Schools in 1979.
Sorry, that was unthinking of me... I'm so used to just typing Sys V when
talking about AT&T unix.
> > Microsoft paid royalties to AT&T.
>
> Correct. And Gates didn't want to pay royalties on the product
> it sold to IBM. Conversely, IBM had just lost a major portion
> of it's Series 1 market to CP/M, and may have figured that UNIX
> on the 8088/8086 was too likely to blow away the IBM Proprietary
> markets mentioned above.
I don't recall MS ever selling Xenix to IBM.
> I was programming for Series 1 (EDX 2.x and EDX 3.x) and CP/M
> machines when the first IBM PCs came out. They were expensive,
> they didn't run much software, and they didn't support hard drives.
> Even the floppy drive was an expensive option (but a popular one,
> thank goodness).
The IBM Winchester driver was one of the first hard drives for personal
computers. IBM PC's had BASIC in ROM, so they often didn't need a disk
drive for early applications. 123 was the magic bullet though.
> Back in those days, IBM and Microsoft wanted you to jump right into
> the middle of BASIC-IN-ROM. Much of this was to be certain that
> CP/M-86 didn't start taking back the IBM market.
No, I think it was just because IBM didn't invest much money into the PC
architecture at first.
> > Lots of people have NT source code liscenses.
> > Bristol, Mainsoft, Compaq,
> > IBM, the US government...
>
> How many of those licenses are current?
>
> How many people get to look at this source code?
>
> How much does it cost?
>
> What are the legal consequences of looking?
>
> (I know many of these answers, but you're the one making the case).
All that is irrelevant. Your argument was lack of source code. My argument
is that it's available if you absolutely need it. Price is not an issue for
a company like AT&T.
> > BTW, gates never said he found the source to BASIC in the dumpster.
> > IIRC, he said he fished out OS memory dumps.
>
> He mentioned both in the interview. Today he's very quiet about
> these things. It came back recently when Microsoft tried to claim
> that Oracle was illegally spying on them by having people go through
> Microsoft's trash.
Not that I ever read. The only statements I saw in the interview discussed
OS memory dumps.
> The problem was that 100,000 lines of code printed on green-bar
> fan-fold paper on a Dec line-printer weighed about 20 pounds and
> was very difficult to shred (since it had to be broken into 2-3 dozen
> shredder size chunks.
No basic interpreter was 100,000 lines of code, especially not back in the
early 1970's.
> As for the IA64 port, even NT 2000 doesn't support it. This hasn't
> made Intel happy since Linux has been creating markets for the
> 64 bit ALPHA and the 64 bit PPC (G5?) and the UltraSPArC. And
> Pushing people into UNIX opens up a number of chip/bus alternatives.
Where do you get this information? It seems counter to the published press
releases from both MS and Intel which claim that 64-bit Windows 2000 will
ship Itanium. MS also announced Windows 2000 (and demo'd it) running on
Itanium the same day that the first hardware was made available.
Additionally, a beta of 64 bit Windows 2000 for Itanium is available if you
want it. Sounds pretty amazing for an OS that doesn't support it.
http://www.microsoft.com/PressPass/press/2000/Feb00/MSWin64Pr.asp
http://www.microsoft.com/PressPass/press/2000/Jul00/ItaniumPR.asp
Also, this article talks about Intel showing both Linux and Windows 2000
running on Merced prototypes in August of 99:
http://news.cnet.com/news/0-1003-200-120832.html?tag=st.ne.1.srchres.ni
> Microsoft currently plans a 64 bit implementation of "whatever"
> around 2001 (which means delivery in 2002?).
Not according to Microsoft:
http://news.cnet.com/news/0-1003-200-2243260.html?tag=st.ne.1.srchres.ni
"Two final versions of 64-bit Windows, one for servers and one for
workstations, will be released when computer manufacturers begin selling
their systems, said Michael Stephenson, lead product manager for the Windows
enterprise server division."
> Meanwhile, the SCO/IBM/HP/DELL/COMPAQ port of UNIX to Itanium,
> plus the availability of Linux 2.4 for Itanium leaves Microsoft
> waiting for the next train.
Considering that Itanium systems aren't shipping to anyone but developers,
and Developers can get Win2000 for Itanium... I fail to see how you can
support that statement.
> > > Attempts to port NT 3.51 to MIPS, Alpha, PPC, and 68000 were so
> > > bad (lack of ISV support) that NT 4.0 was only ported to the ALPHA.
> > > And now, Windows 2000 isn't even available on the ALPHA.
> >
> > More lack of knowledge on your part Rex. The NT4 CD shipped with
> MIPS,
> > ALPHA, and PPC as well as x86. 3.51 didn't have an Alpha port, and
> > considering that NT was *DEVELOPED* on MIPS and ported to x86, it
> makes your
> > statement even more ludicrous.
>
> But there was no market level support.
Not the same thing as what you said. You claimed NT 4.0 was ported only to
Alpha. Which is completely untrue.
> Many of these companies spent
> over $1 billion on the NT ports, only to have Microsoft fail to provide
> even Office 2000 for their chips. Ironically, Microsoft provided IE 4.0
> for UNIX versions but not for NT/MIPS or NT/PPC or NT/ALPHA.
> Appearantly,
> they were supposed to fork out $1 billion each for any application they
> wanted.
MS didn't see any market for Office on non-x86 Win32, and they were probably
right. Most of thse systems were either servers, or specialized
workstations. Though it was a bit of a chicken and egg scenario.
> I had heard that the NT 3.x development platforms included UNIX
> workstations from SGI, since they needed workstations that could
> handle/compile/manage the NT code (Windows 3.1 couldn't). I didn't
> know that they had done the original port to MIPS machine level
> instructions (do you have a reference?).
Sure, read Inside Windows NT (either the original 1992 Helen Custer Edition
or last years heavily revised 2nd edition). But if you must...
http://www.microsoft.com/WINDOWS2000/news/fromms/kanoarchitect.asp
""We certainly achieved extensibility and portability," Lucovsky says. "We
tested ourselves by not doing the x86 version first. We did the RISC
(Reduced Instruction Set Computing) stuff first. It would have been so easy
to drop the RISC support; everyone in the company wanted to. But the only
way to achieve portability is to develop for more than one platform at a
time. It cost us a lot to keep portability alive, but we did, and that has
made it easy for us to respond to things like Merced," he says, referring to
the 64-bit chip from Intel."
BTW, you might be interested in this article:
http://msdn.microsoft.com/library/winresource/ssreskit/rk_ntsvrunix_sjuq.htm
Windows NT from a Unix point of view.
> You certainly do seem to have very intimate knowledge of the internal
> workings of Microsoft. Where do you get this interesting information?
> Got a history book I can browse through? (so I can "refresh my memory").
I followed the industry buzz for years. I've also read lots of interviews
and books. (Check out "Barbarians led by bill gates" for instance).
> > > They do, they have, and for enough of an incentive (like a huge
> > > percentage (say 20%) of your company) they might do it again.
> > > Even Paul Allen is betting against Microsoft (Transmeta, LinuxWorld,
> > > and other Linux Friendly ZDNet publications). And he's dropping
> > > Microsoft stock like soap in a cold shower.
> >
> > Sorry, Microsoft SEC filings do not show that Microsoft owns 20%
> > of the US government, Mainsoft, or Bristol.
>
> Microsoft takes smaller shares of larger companies and larger shares
> of smaller companies. Furthermore, Bill Gates, Paul Allen, and Steve
> Ballmer all have secondary holding and VC companies. Also, Microsoft's
> top three institutional investors DO own substantial portions of
> some of these organizations.
Give me a break. Bill Gates, Paul Allen, and Steve Balmers personal
investment portfolios have little if anything to do with Microsoft source
code liscenses. It would be an extremely huge SEC violation for Bill Gates
to personally be granted shares of a company in excange for favorable
liscensing. This stuff is all very well documented. So please provide some
references.
> Although Microsoft doesn't own 20% of the U.S. government, they
> get nearly 5% of the GNP as revenue (30 billion on 6 trillion)
> and have repeatedly been granted extraordinary favors from the
> U.S. government.
>
> Even the DOJ's efforts to consolidate 22 independent lawsuits into
> a single case was at the request of Microsoft. If Microsoft had
> known that the 22 prosecutors were going to hold a poker to the
> DOJ prosecutor's back, they might have been less interested in
> a consolidation. Now they are trying to "take it back" by
> challenging jurisdictional rights (essentially forcing each state
> to persue it's claims independently).
Such things are commonplace. The tobaco industry also got granted similar
"favors". It costs money for the government to sue someone. It's deemed in
the best interest of the people to consolidate such suits.
BTW, the jurisdictional challenge is only relating to the fast track supreme
court appeal on the grounds that state cases cannot bypass the appeals
court. MS isn't trying to take anything "back" but rather play under the
rules established.
> > > And by 1995, there was another new company - Caldera. It was formed
> > > by some of the Novell people who were upset that Novell had agreed
> > > to cancel it's workstation initiative if Microsoft agreed not to
> > > deploy NT as a server.
> >
> > Considering that MS had deployed NT as a server 3 *YEARS*
> > before Caldera was formed, that seems highly silly.
>
> However, it shows that there were people within Novell who were
> actively supporting Linux (especially as a workstation platform).
> I got NT 3.5 beta in 1983, and NT 3.51 in late 1983. When did you
> get yours?
Uhh.. you mean 1993. NT 3.51 didn't go into beta until 1994. 3.51 was
really just 3.5 with the Windows 95 common controls and Win32c (the chicago
API's) grafted on (but mostly stubbed out or redirected to other API's).
> Most of the PCI PnP drivers came out for Win95, not NT. NT had a hard
> time getting drivers even AFTER it went GA.
Yes, that's true (Finally).
> > > Keep in mind this was NT 3.5/3.51. Some of those folks were backing
> > > Linux as a workstation platform back in 1994.
> >
> > NT 3.1 had a server version.
>
> Exactly when did NT 3.1 go GA?
July 1993.
> I remember a BETA version that was so buggy it was unreliable
> even as a file-server. Was that back in late 1992?
> Dow Jones wasn't even allowed to publish anything (except
> glowing reviews approved by Microsoft's editorial department).
3.1 was pretty stable overall, but very slow. NDA's generally disallow
publishing of such information, since beta's are not intended to be stable.
> > > > ???? Lower level MFC???? Gezus Rex,
> > > > stop while you've got SOME credibility.
> > > > MFC is not an API in any sense of
> > > > Windows (it's a class framework) and
> > > > there's no such thing as a "low level MFC API"
> > > > even if you're stretch the
> > > > terminology to loosely accept MFC
> > > > as some sort of API at all.
> > >
> > > I have some old C++ manuals that were published in 1992 that would
> > > be quite a bit different from anything you've read.
> >
> > I was developing on C++ in 1991. Sorry.
>
> [quoting you below]
> > Visual C++ version 2 included no API's other than direct Windows and
> > MFC.
> [end quote]
I said "MFC is not an API in any sense of WINDOWS". I didn't say it wasn't
an API. It's just not a Windows API (other than the loosest of definitions
which say "It's an API and it runs on Windows, so it's a Windows API". The
same could be said of Corba).
> I have 20 years of programming to both Microsoft and UNIX platforms,
> but you seem to have a better "History book" than I do.
I have a better memory than you do.
> Since you never post your URL, I have no way of establishing a common
> reference point. You seem to know what you're talking about, and
> you like to pick nits - I'll deferr to your undocumented "expert
> opinion".
I don't call correcting gross misapplications of history, massively
incorrect statements, and wild flights of fantasy "picking nits".
> Perhaps you'd like to give your own dissertation on why NT 3.1 through
> Windows NT 3.51 were so unstable.
NT 3.51 was not unstable at all. In fact, it was so much more stable than
NT4 in general that MS ran it's own web servers on 3.51 until Windows 2000
came out.
NT4 had many instability problems relating to certain services (at least up
until about SP3).
> Perhapes you'd like to give your own dissertation on why Windows 2000
> is soo much more stable than Windows NT 4.0 and Windows 95.
Because MS developed and (more importantly) USED a comprehensive design and
development methodology for Windows 2000. They literally invested billions
of man hours into the re-design and re-implementation of much of the core NT
architecture. It's truly amazing that MS pulled off writing 20 million
lines of new code and rewriting much of the original 15, and came out with
something orders of magnitude more stable than the last release.
> > > > CORBA Client GUI objects... CORBA didn't even have a component
> > > > architecture until CORBA 3, which was very recently.
>
> True, but many component architectures were based on CORBA, including
> GNOME, KDE, GTK, and Qt.
GTK and Qt do not use CORBA at all. They're just X toolkits. GNOME uses
CORBA, but I don't think KDE does (I could be wrong on that).
> > > > > Many of the features of COM, such as drag-and-drop, dde, and
> > > > > embedding are not supported for X11 interfaces.
> > > >
> > > > Because those are desktop features, not distributed features.
> > > > They don't work in DCOM at all, even between windows machines.
>
> However, if a DCOM server object wants to invoke many of methods
> that are required on the Client side, these clients are not supported
> on UNIX.
Wrong. Unix DCOM supports everything that Windows DCOM does. Again, the
features you mention are *COM* interfaces, not DCOM interfaces. They are
strictly prohibited from DCOM because they use OS dependant (and machine
dependant) data such as window handles, or memory contexts.
> Eventually, the OMG just absorbed a DCOM <-> CORBA Mapping scheme
> which makes it possible to invoke a CORBA object as a DCOM object
> and vice-versa.
But guess what? This "bridge" requires both a DCOM and CORBA implementation
to work.
> My references are below - where are yours?
Your "references" are just more statements made by you. That's like an
author using his own text in his bibliography.
There is no proof that much of anything you say there is true.
------------------------------
From: "Slava Pestov" <[EMAIL PROTECTED]>
Subject: Re: Malloy digest, volume 2576+8695i
Crossposted-To:
comp.sys.mac.advocacy,comp.os.os2.advocacy,comp.os.ms-windows.nt.advocacy
Date: Thu, 24 Aug 2000 16:32:09 +1000
In article <LFWo5.341$[EMAIL PROTECTED]>, "Joe Malloy"
<[EMAIL PROTECTED]> wrote:
> Nor did he explain why he's assuming Slava's question
No explanation was necessary on Tholen's part, Malloy. Meanwhile, you
still haven't answered my question, Malloy.
> [who is this "Slava," Tholen,
Don't you know? Meanwhile, you still haven't answered my question,
Malloy.
> one of your sock puppets?].
Incorrect. More evidence of your illogic. Meanwhile, you still haven't
answered my question, Malloy.
------------------------------
From: "Erik Funkenbusch" <[EMAIL PROTECTED]>
Crossposted-To:
comp.os.ms-windows.nt.advocacy,comp.os.os2.advocacy,comp.sys.mac.advocacy
Subject: Re: Would a M$ Voluntary Split Save It?
Date: Thu, 24 Aug 2000 02:14:44 -0500
"R.E.Ballard ( Rex Ballard )" <[EMAIL PROTECTED]> wrote in message
news:8o1shh$ea$[EMAIL PROTECTED]...
> > Software doesn't do this, so companies
> > need to continually come up with new
> > ways to get customers to buy new versions,
> > and that's typically to enhance
> > the product and make it better.
>
> Microsoft does this one better. It deliberately breaks
> backward compatibility.
>
> Try running Office 2.0 on Windows NT.
There was no such thing as "Office 2.0" The first version of Office was
4.0.
But, Word 2 and Excel 3 (or was it 4?) Run fine under NT. These would be
analagous to an "office 2.0" if it had existed.
> Try running Office 2000 on Windows 3.1
That's silly. Try running native PPC applications on a 68000 Macintosh.
> Try reading Offire 2000 documents with Office 95.
No problem for Word and Excel (you do have to download the Word 2000 filters
from MS though). The Excel format hasn't changed much since Excel 5.
http://officeupdate.microsoft.com/downloadDetails/wd97cnv.htm?s=/downloadCat
alog/dldWord.asp
I can't speak for PowerPoint, and Access is an entirely different beast.
> It's a very effective business strategy if customers are ready,
> willing, and able to pay for hardware and software upgrades
> every 3 to 12 months (required for any organization of over
> 1000 users).
Or just get the new filters.
> That Windows 3.1 machine was an 80386/33 with 4 meg of RAM and
> a 40 meg hard drive. It included Word, Excel, and Power-point (more
> like Draw), which hadn't been packaged officially as "Office" yet.
> It crashed several times a day when in regular and frequent use,
> and the only way to get it to run reliably was to run each application
> in full-screen mode. The biggest problem was that if you put a window
> in the background and it tried to publish an OK/Cancel dialogue, it
> didn't display the dialogue where it could be reached (since it was
> covered by the forground application). But since it took exclusive
> focus of the keyboard and mouse, the only way out was to reboot the
> machine.
Think about this for a moment. If the dialog took keyboard and mouse focus,
then that means it was accepting keyboard input. A simple enter key or
escape key would get out of it.
I never saw anything like this happen.
> It's competitor, at nearly twice the price, was SunOS 4.0, which
> included a 21 inch monitor, an optical mouse, a 500 meg drive, and
> a new application suite called Applix Asterix. It invited the user
> to create lots of windows, overlap them with impunity, and even to
> position them so that real-time displays such as strip-charts,
> scrolling status screens, and other real-time feedback was available
> to the user. The user could cut-paste, using X11, from application
> to application with very little required from the application.
>
> It was expensive, but it worked.
I would hope so.
> 10 years later, Microsoft has gone through 4-5 replacement level
> revisions, including Windows 3.1, Windows 3.11, Windows NT 3.51,
> Windows 95, Windows NT 4.2, Windows 98, and now Windows 2000.
> Fortunately, many of us skipped one or two releases (not me of
> course, I bloodied myself on everything but Windows 98).
NT 4.2???? Come on Rex. Even *I* know you know there's no such thing.
> Meanwhile, back at the ranch, those old Sun SPArCs are still working,
> many of them are now working as servers at ISP locations, as
> application servers, or even as file & print servers. Although
> some of the newest Solaris stuff doesn't run on it, the Linux software
> ports very nicely. And those 10 year-old applications still produce
> formats that can be read/published to/from everything from Asterix
> to Office 2000 and the very latest versions of WordPerfect, WordPro,
> Words, and StarOffice Word.
And those old Windows PC no longer work? Windows magically stopped working
just because new versions came out? Applications miraculously stopped
functioning just because time went on?
> 10 years is ancient for a computer, but there are still SPArC 10s
> and 20s being sold on e-bay. They're like the eveready bunny,
> they keep going and going and going...
And there's still 286's being sold on ebay. So what?
> > Whether MS has competition from other companies or not,
> > it is still in competition with all the copies of the
> > software it sold before.
>
> This is very true. I'm still running Office 97, NT 4.0, Windows 95,
> and IE 3.0. This is because I ALSO use Mandrake 7.0, Netscape 4.7x,
> and Applix Office 5.0
>
> You see, the upgrades are such a hassle, that the risk of losing
> critical information hidden in some system directory, that will
> cost me many hours/days/weeks to recover, at $300 an hour, is just
> such a hassle, that I'd rather NOT hassle with the upgrades.
First you claim you've upgraded to everything, now you claim you don't?
> Sure, I tried Windows 98, but it choked on my older PC. I tried
> Windows 2000 but it choked on my older Laptop.
You just said you didn't try Windows 98. Which is it?
> Here's the joke. In the 10 years since I first did that survey,
> the average Windows user spent nearly 5 times what the average UNIX
> user spent on software and hardware. The UNIX user was more productive,
> generate more revenue for his company (through the Web, application
> services, high capacity databases, and complex applications). The
> Linux users are off the scale. They spent less, made more, and
> reached a higher level of competency than either of their compatriots.
> Much of this was out of necessity (Slackware 1.1 was VERY ROUGH).
You are forgetting about the lack of critical vertical applications for
Unix. Many companies run some kind of special purpose application relevant
to their industry, which often only exists on one platform.
> I not only see that Applications will improve for Linux, but that
> we will see a whole new Genre of killer-apps for Linux that will
> have capabilities that would make Microsoft jealous. Things like
> real-time feedback, real-time animated charting (something like
> performance monitor, but with 3D rendering, real-time charting,
> logging, analysis, summaries, smoothing, and histograms. This could
> be used to monitor everything from stock prices to cash-flow.
Let me know when this happens.
> Imagine the first TV interview where the CEO of a huge company like
> Federal Express is sitting in an interview, is asked how his company
> is doing at the moment, pulls out his Ipaq/Linux hand-held with
> wireless modem, hits a button, and you can see a second-by-second
> of target verses actual revenue, expenses, and earnings.
So you do live in a fantasy world. This isn't likely to happen. Ever.
Most CEO's I've met (especially from huge companies) don't even know how to
turn on their PC. They have assistants to do that stuff. I'll bet Michael
Eisner doesn't even have a PC in his office, much less a laptop or handheld.
------------------------------
** 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
******************************
