Linux-Advocacy Digest #451, Volume #31 Sun, 14 Jan 01 10:13:02 EST
Contents:
Re: Linux *has* the EDGE! (Karel Jansens)
Re: Windows 2000 (Karel Jansens)
Re: A salutary lesson about open source (Karel Jansens)
----------------------------------------------------------------------------
From: Karel Jansens <[EMAIL PROTECTED]>
Subject: Re: Linux *has* the EDGE!
Date: Sun, 14 Jan 2001 15:37:04 -0100
You might want to check
http://www.pla-netx.com/linebackn/guis/win1983.html
for information on Windows 1.0.
Particularly interesting is the text under photo 8 in the review which
says:
"Photo 8 shows both the Multiplan window and the Microsoft Word window
reduced. (Since photo 8 was taken, Microsoft Windows has been adapted to
use an automatic resizing process called "tiling." Rather than letting
windows overlap or leaving part of the desktop empty, Microsoft Windows
always gives all the space on the screen to the applications that are
running.)"
This indeed indicates that Windows 1.0 did not have overlapping windows
(and in true Microsoft philosophy, this was deemed a feature).
In honesty though, I have to admit that I never encountered the problems
with overlapping windows on 3.0 and 3.1 you describe in this post.
Regards,
Karel Jansens
"R.E.Ballard ( Rex Ballard )" wrote:
>
> In article <pEX76.1251$[EMAIL PROTECTED]>,
> "Erik Funkenbusch" <[EMAIL PROTECTED]> wrote:
> > "R.E.Ballard ( Rex Ballard )" <[EMAIL PROTECTED]> wrote in message
> > news:93p97a$73h$[EMAIL PROTECTED]...
> > > > > Windows 3.1 used cooperative multitasking.
> > >
> > > Correct. And SunOS and Linux and FreeBSD and SCO all
> > > provided real-time multitasking AND excellent support for
> > > multiple overlapping windows, back when Windows 3.0 was
> > > still trying to get overlapping windows to work. It was
> > > marginal at best.
> >
> > Hmmm.. given that Windows 3.0 had
> > overlapping windows on the day it was
> > released, and FreeBSD didn't exist when
> > Windows 3.0 was released, that would
> > be quite a feat.
>
> Correct. I meant BSD, not FreeBSD. Windows 3.0
> had overlapping windows, but there were severe
> problems when you actually used them. At the time
> that Windows 3.0 was released, the big competitors
> were "Software Carusels" that let users switch
> from application to applications in full screen mode.
>
> Even Windows 3.1 was usually used in "Full Screen" mode
> most of the time. I've had numerous conversations with
> people who can't understand why anyone would think
> Windows was unreliable. When I look over their shoulder,
> I notice that they almost only run only one window at a time.
> In many cases, they even iconify the windows. I recently
> looked over one person's shoulder and watched as he completely
> stopped and started each application numerous times. I suggested
> that he could run multiple applications and he commented; "If I
> do that it crashes too often". Three weeks later he was swearing
> that Windows was reliable.
>
> Windows 95 has a number of design features that are deliberately designed to
> encourage the use of full-screen windows. Windows NT 4.0 adopts most of
> these user interface features even though it is very good and fully
> functional
>
> > I think you're confusing Windows 1.0 (which didn't
> > have overlapping windows) with Windows 3.0, and
> > confusing FreeBSD with BSD.
>
> Windows 1.0 didn't have overlapping windows AT ALL. Windows 3.0
> had them, but if you used them your PC crashed very regularly,
> often within an hour.
>
> > > > > It meant applications had to
> > > > > yield to achieve multitasking.
> > >
> > > Yup It's called "busy/wait" multitasking.
> > > It was obsolete back in the early
> > > 1980s, but Microsoft revived it for Windows.
> >
> > Actually, Apple started it.
>
> Actually, Busy/Wait was first used in FORTH. Even today,
> many FORTH systems are used for robotics and engine control
> on automobiles, as well as thermostats, microwaves, and VCRs.
>
> In fact the most widely used operating environment used in the world isn't
> Windows, it's FORTH, which is both a language, an Operating system, and an
> application environment. Most housholds have at least 5-7 forth engines in
> their house (VCR, TV, Microwave, Automobile, Thermostat).
>
> > > > > Windows
> > > > > 95/98/ME/NT/2000 are all pre-emptive multitasking.
> > >
> > > True but the multitasking is still sigificantly inferior to
> > > Linux or UNIX. Windows 2000 was substantially improved, but
> > > all of the applications would have to be redisigned and
> > > reimplemented
> > > to exploit the improvements.
> >
> > What are you talking about?
> > You don't have to redesign any apps to get good
> > multitasking in Win2k (unless you're talking SMP,
> > which is not the same thing).
>
> To get the full benefits of memory protection and rapid context switching
> from Windows 2000, applications should be modified to support MTS, Fabrics,
> and Pooling. The technology is actually quite exciting, since it's a
> significant improvement in reliability.
>
> The old NT 4.0 and Windows 95 applications will usually run an
> Windows 2000, as long as you aren't using VBX at the application
> levels.
>
> > > 3rd party developers can invest millions
> > > trying to accomodate the limitations
> > > of the various versions of Windows,
> > > or they can leverage the core structures
> > > provided by Linux and UNIX.
>
> > > But there isn't much profit in Windows.
> >
> > Well, I guess that explains why Intuit
>
> Intuit was protected by a federal court order
> that prevented it from being
>
> > and AOL are multi-billion dollar companies.
>
> Microsoft tried to kill AOL and failed. AOL purchased
> Netscape and Microsoft threatened to choke AOL out of
> the Market if they made any attempt to improve or promote
> Netscape Navigator themselves.
>
> There are a handful of companies who have done well under
> Microsoft, usually because Microsoft owns controlling interest.
>
> But for each one of those, there are dozens of companies who thrived and were
> destroyed by Microsodft. Companies like Borland, Novell, WordPerfect, Lotus,
> Corel, DBase, Real Networks, Lattice, Stacker, ButtonWare, Trumpet,
> Chameleon.
>
> > > Star Office is slow but tolarable.
> > > Most of the applications that are unique
> > > to Windows run under WINE, and what's left runs under VMWare.
> >
> > Most? Hardly any. the Wine status page
> > lists only a 30-50% CRT compatibility,
> > that means that any apps that use the
> > MSVCRT are 50% or more likely to die.
>
> Lotus Notes 4.5 and 5.0 run quite nicely under WINE.
> So does Corel Draw. Project 95 also seems to work O.K.
>
> Linux doesn't have to support every Windows application,
> just certain strategic ones.
>
> > > Windows 2000 isn't bad, but I'm waiting for either SP2 or SE.
> > > Windows 2000 just isn't worth $300.
> >
> > The upgrade from Win9x is much cheaper.
>
> True. Today, Windows 2000 upgrade is about $189. That is for
> the workstation version, at Staples or Comp USA.
>
> > > > > If you're talking about Linux and the CLI,
> > > > > I would agree
> > >
> > > CLI, Scripts, the ability to quickly script
> > > applications and put GUI
> > > front-ends on them to launch them.
> > > In fact numerous front ends.
> >
> > You do realize that Tcl/Tk exists for windows, right?
>
> Absolutely. Which is exactly my point. Why code software
> in "Windows Exclusive" languages like Visual Basic or C# when
> you can just as easily use PERL, TCL, PYTHON, and JAVA? All
> of which let you kill two birds with one stone.
>
> Why use ASP on IIS when you can run PHP on Apache and run on either Windows
> 2000 or UNIX/Linux?
>
> Why use Active Directory when you can use Kosher LDAP and Kerboros
> (available in source form as SLAPD) or something that really is
> LDAP compatible.
>
> I'm suprised that the FTC hasn't fined Microsoft for
> false advertizing infractions. Active Directory is
> as much like LDAP as IPX is like IP. There are
> similarities, and Microsoft clearly borrowed the
> specifications, but Microsoft has added fields which
> are neither documented, nor do they conform to any
> accepted or proposed IETF standard. Because the system
> fails to function with other LDAP and Kerberos systems,
> advertizing Active Directory as LDAP or Kerberos is
> both misleading and fraudulent.
>
> > > Windows 2000 has lots of Eye Candy and
> > > really snappy displays (assuming you
> > > have enough RAM and Video RAM to support it.
> >
> > how fast your displays are have nothing
> > to do with how much video RAM you
> > have (except for 3D).
>
> I've used a couple of different video cards, and it seems like
> the cards with more VRAM seem to refresh more rapidly. This
> could just be my own perception.
>
> > > > > > Windows 2000 has a nasty habit of getting
> > > > > > into a state where the only
> > > > > > way to fix it is to reinstall the software.
> > > > > > In some cases, it won't
> > > > > > even reboot. Fortunately, this only happens
> > > > > > every 8-10 weeks, but it's
> > > > > > still very annoying.
> > > > >
> > > > > I've not seen this.
> > >
> > > I've had it happen about 4 times since the first production releas
> > > (which I bought 2 weeks after the announcement.
> >
> > You just said you were waiting for SP2 to buy 2000.
>
> I'm waiting for SP2 before I try yet another installation. I've
> already paid the money. I upgraded one of my NT 4 machines, and
> have used Windows 2000 enough to see that it's better than anything
> else Microsoft has to offer. Actually, I'm surprised that Microsoft
> didn't just push the OEMs into a consumer version of Windows 2000.
>
> Windows 2000 has a number of technological "anti-linux" measures. It took a
> bit of time for the Linux community to figure out work-arounds. I personally
> love that Windows 2000 supports both FAT 32 and NTFS.
>
> Of all Microsoft offerings, I like Windows 2000 the best.
>
> I just like Linux better.
>
> I think many people, but not all, will also like Linux better when
> they are able to make the choice.
>
> It would actually be in microsoft's best interest to just cooperate
> with Linux rather than trying to force an "all or nothing" solution.
> If Microsoft pushes too hard, they may even find that the market
> for Windows 95 (pirate or legal) will kill BOTH Windows 2000 and
> ME.
>
> Furthermore, if Microsoft torpedos too many 3rd party applications,
> those companies are likely to put more support into WINE, just as
> Corel and Lotus did.
>
> > > This may partly be because Windows deals with
> > > the entire result set as a single object. UNIX deals
> > > with the result as a stream of results. As a result,
> > > you can begin displaying what might be a result set
> > > of several thousand files (remember back in the days
> > > with FAT 16 would only let you have 256 files per
> > > directory). When you have to treat the entire response
> > > as a single object, you must allocate memory for the entire
> > > object, and allocate process space and kernel space for
> > > the entire object. When the objects are huge, the situation
> > > can get pretty ugly.
> >
> > You seem to have a knack for claiming
> > the same thing three times and making
> > it look like 3x the problem.
>
> > Allocating memory is the same thing
> > as allcoating process space. kernel
> > space is not allocated for anything but drivers.
>
> I'm focusing on three different aspects of memory management under Windows.
> Windows has memory management problems. This includes Windows 2000.
> Fortunately, Microsoft resolved 2 serious problems in Windows 2000. First,
> the libraries were restructured to reduce the amount of code thrashing
> (something I suggested in this forum back in 1998). They also organized the
> buffer space more efficiently to make "copy on write" more efficient.
> Microsoft offered me a job as their way of showing their thanks, but withdrew
> the offer when they realised "I didn't have the Microsoft religeon". It's
> really not that bad. I would have had to take an 15% cut and the Options
> were at $120 a share and have dropped to $41 a share since then. At $41 a
> share, even I found Microsoft stock to be attractive.
>
> > > I think we agree that GUI preferences are a matter
> > > of personal preference. Windows 2000 gives you ONE option.
>
> > > Linux just happens to give you your
> > > choice of about 6 (more if you count the styles).
> > >
> > > Microsoft wants you to ues VB because you can't
> > > put it elsewhere. Microsoft doesn't want you to
> > > use PERL, JAVA, TCL, or any other "cross platform"
> > > environment, because you might stop paying 10 times
> > > the production cost for Microsoft products.
>
> > > And Microsoft wouldn't be able to get 50% profit on
> > > revenue.
> >
> > MS doesn't "want you to use VB".
>
> Oh yes they do. They spend nearly $1 billion in advertising
> carefully structured to motivate coverage of VB and to prevent
> coverage of PERL, TCL, and Python.
>
> > They make VB because it's what their customers want.
>
> I would challenge that. Microsoft created incentives for using VB,
> and penalties for not using it. In some cases, they even witheld
> APIs from other languages made them much more complicated do
> discourage the use of portable languages.
>
> Again, Microsoft is protecting their control of the desktop market,
> with hopes of extending that to other markets.
>
> > MS makes several languages, including Visual Foxpro (which
> > also can't go anywhere else, yet they don't push that).
> > Besides, things are changing with .NET.
>
> FoxPro was pretty much rendered obsolete by Microsoft Access and
> SQL Server. If you need a multiuser server, SQL Server is better
> than FoxPro. If you just need a "quick and dirty" database, Access
> is bundled with Office.
>
> FoxPro isn't a Microsoft strategic project. It's more like a legacy
> product which Microsoft keeps alive to support customers who would
> switch to other products if Microsoft was to drop FoxPro.
>
> > In a years time, we'll see MS applications runing
> > unchanged on MacOS X, Palm, Win CE, Itanium, and x86.
> > Perhaps even Solaris or HP/UX.
>
> But not Linux?
>
> This is why the Linux and UNIX community doesn't trust any of Microsoft's
> tools and protocols. The fact is that Microsoft has made a number of
> contributions to the Linux and UNIX community, but primarily to protect their
> markets. For example, when the Open Systems community threatened to enhance
> RARP to provide automatic address configuration, Microsoft published DHCP
> assured that Microsoft clients would be compatible with UNIX servers (which
> were replacing outgrown Windows NT 3.51 servers).
>
> SMB and WNS were published when GPL file systems such as NFS and AFS
> were about to be offered for Windows.
>
> Microsoft tried to lock Linux out of the internet using MS-CHAP,
> but leaked the information when Linux servers threatened to lock
> out Microsoft MSN customers.
>
> The open source community has to literally give stuff away to
> get Microsoft to give away even minimal standards.
>
> In many cases, the open source community even helps close security
> holes created by these standards.
>
> --
> Rex Ballard - Sr I/T Systems Architect
> Linux Advocate, Internet Pioneer
> http://www.open4success.com
> Linux - 60 million satisfied users worldwide
> and growing at over 9%/month! (recalibrated 10/23/00)
>
> Sent via Deja.com
> http://www.deja.com/
------------------------------
From: Karel Jansens <[EMAIL PROTECTED]>
Subject: Re: Windows 2000
Date: Sun, 14 Jan 2001 15:38:37 -0100
"T. Max Devlin" wrote:
>
> Said Joseph T. Adams in comp.os.linux.advocacy on 14 Jan 2001 00:31:19
> >Russ Lyttle <[EMAIL PROTECTED]> wrote:
> >: The real truth is that Excel for the PC is so tied to the PC that MS
> >: couldn't port it. So they wrote another product that had a UI similar to
> >: Excel and called it Excel even though it isn't.
> >
> >
> >Actually, Excel for the Mac predated Excel for Windows, and for that
> >matter Windows itself.
>
> Well, not 'Windows itself'. Windows 1.0 (a side-kick like
> task-switching GUI desktop) predated Excel, I think. Excel for Windows,
> however, predates Windows 3.1, the first Windows that anyone really
> bothered with.
>
1. Is it really fair to compare Windows 1 to such a fine product as
Borland's Sidekick?
2. True. Just the other day I happened to be leafing through some 1991
PC rags and in one of them was a review of Excel for Windows 3.0 (I
think it was in PC Laptop - Is that still around BTW?). The reviewer's
biggest complaint was that the program took a whopping 5 MB of disk
space and required at least 2 megs of RAM to run comfortably.(*)
BBTW, I recommend anyone to re-read some of those prehistoric
publications; it really does put things into perspective.
Regards,
Karel Jansens
(*) So I guess today's machines must be at least 32 times as productive
then...
------------------------------
From: Karel Jansens <[EMAIL PROTECTED]>
Crossposted-To: comp.os.ms-windows.nt.advocacy
Subject: Re: A salutary lesson about open source
Date: Sun, 14 Jan 2001 16:07:17 -0100
Chad Myers wrote:
>
> "Jim Richardson" <[EMAIL PROTECTED]> wrote in message
> news:[EMAIL PROTECTED]...
> > On Fri, 12 Jan 2001 21:29:02 GMT,
> > Chad Myers, in the persona of <[EMAIL PROTECTED]>,
> > brought forth the following words...:
> >
> > >
> > >Who said that? Not me.
> > >
> > >It's funny, you guys say, "Open source is superior"
> > >I say, "No it's not, look at X"
> > >You say, "Oh, so closed source is perfect, right!?"
> > >
> > >Um... no, I'm saying Open source isn't superior, nor perfect, nor
> > >anything the OSS advocates claim it to be. It's no better, only
> > >worse than closed source.
> > >
> >
> >
> > Has it been pointed out to you that it took 6 months as open source, to
> > discover a backdoor that had existed in a previously closed source program for
> > years?
>
> No one was looking because no one needed to.
>
Why not?
> how was closed source better in this case?
>
It wasn't, is what you'll probably find the point to be.
> How many times had it beel exploited?
>
You can't know, can you?
Doesn't that scare you?
Shouldn't it?
> What about the people who now have the closed source version, and haven't
> upgraded to the open source version who's exploit has now been made public
> knowledge.
>
If they are stupid enough to continue using something that has a proven
and documented security risk, they deserve anything that's coming to
them.
> > If it was still closed source, the backdoor would still be there, and we
> > would not know about it.
>
> But no one knew about it, there were no cases of exploitation. Now
> people know about it, and there are customers who are now sitting ducks
> thanks to Open Source.
>
Your logic is flawed: there was at least one person who knew about the
backdoor, the one who put it in in the first place.
There are no _known_ cases of exploitation, which can mean that either
there have been none, or that the victims chose not to make it public.
It's not as if that has never happened before.
Any "sitting ducks" do indeed share the intelligence level of their
wooden namesakes.
Regards,
Karel Jansens
------------------------------
** 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
******************************
