Linux-Advocacy Digest #300, Volume #34 Mon, 7 May 01 15:13:04 EDT
Contents:
Re: bank switches from using NT 4 ("David Brown")
Re: Drive Partitions (Dave Berntson)
Re: bank switches from using NT 4 ("David Brown")
Re: bank switches from using NT 4 ("David Brown")
Re: How to hack with a crash, another Microsoft "feature" ("Erik Funkenbusch")
Re: Why Linux Is no threat to Windows domination of the desktop
Re: Justice Department LOVES Microsoft! (Bob Hauck)
----------------------------------------------------------------------------
From: "David Brown" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.ms-windows.nt.advocacy
Subject: Re: bank switches from using NT 4
Date: Mon, 7 May 2001 20:36:55 +0200
Jan Johanson wrote in message <3af69816$0$12207$[EMAIL PROTECTED]>...
>
>"David Brown" <[EMAIL PROTECTED]> wrote in message
>> Dropping DOS and moving to a single code base is without doubt a big step
>> forward for MS, but you make it sound like they have just invented sliced
>> bread - they should have taken this step 10 years ago when the 386 became
>> popular. There is nothing magical or dramatic about dropping DOS - no
>other
>> OS has been hobbled by such legacy.
>
>AHHHh you said it right there in that last sentence.
>
There is a BIG difference between backwards compatibility and using out of
date code. When diesel trains were becoming popular, it was important that
their were backwards compatible (same wheel guage) with steam engines - it
did not mean that they had to work by burning diesel to boil water instead
of using wood or coal.
NT is a different kernel from DOS+Windows - it is not legacy, yet it is
mostly backwards compatible.
>
>Wanna take a guess why Windows is so popular?
For every two people in this newsgroup, there are at least three answers to
that question :-).
>Backwards compability (I know,
>not 100% so don't quibble on that point) is the thing.
There is no doubt that backwards compatibility is important - crippling a
modern system with ancient shoddy designs is a matter of poor design and
total disregard for the users.
Consider Wine - an open source implementation of the Windows API for use on
unix-like systems. It has been developed, slowly, by a group of volunteers
with no information about the internals of Windows, and only the vague
documentation of the API to work with. It is maybe around 80% usable (i.e.,
about 80% of apps are usable, or many apps are 80% useable, or something
like that). It provides a strong backwards compatible layer on top of a
choice of modern systems. With a fraction of the effort MS used to produce
the Win9x family, they could have gone straight to an NT-type kernel (or
anything else) and built a fully (or almost fully) backwards compatible
layer on top of it. The Wine project, and the NT family, show there is
absolutely no technical reasons not to have done this - it is purely a
matter of marketting and milking the consumers for more and more money.
>A guy with a copy of
>Windows 95 and software from 5 years ago will mostly likely still be able
to
>run his stuff when he upgrades to WinXP.
MS has a different view to backwards compatibility from the rest of us.
When I think of backwards compatibility, I want to be able to buy an
application today and use it on an OS from five years ago. I also want to
use a five year old app on today's newest OS. I also want to be able to
swap data between old and new versions of programs. MS fights hard to make
sure that although you get a minimum level of backwards compatibility, you
are pushed into upgrading *all* your MS software.
>Hardware from 95 will still most
>likely run in a machine purchased in 2002 (still plenty of motherboards
with
>ISA slots if that's the hang up).
>
>Now - sure, MS could do what you propose. Come out with Windows -3112 "The
>Antiupgrade" and have it rewritten from the ground up by european teens on
X
>supporting only the very latest popular standards and just ignoring
>everything that is "uncool" in the scene at that minute and just ignore the
>VAST installed userbase and blow off every single manufacturer and ISV and
>force them to grok some new undocumented (documentation is uncool and
boring
>to true hackerz) APIs...
>
><sarcasm>yea, great idea!</sarcasm>
>
They could have done as they promised - dropped DOS after Win95. Or maybe
done as they next promised - dropped DOS after Win95 OSR2. Or after Win98.
Or Win98SE. Or Windmill. They have had NT - a mostly backwards compatible
OS family for years, which everyone knows is better in virtually every way
than Win9x, and they have been promising for years that they will drop the
Win9x line and move to the far superior NT kernel. Yet they have
consistently failed to do so. What I propose is that MS, for once in its
existance, actually delivers on a promise and makes XP a single code base.
I also propose that you don't consider it to be something marvelous when
(and if - I'm not holding my breath) they get it right.
>
>>Basically, all they are doing is taking
>> a working non-DOS OS (NT 5.0, aka. W2K) and making it better for games so
>> that home users will want it too. In fact, given that (as far as I
>> understand it), W2K is fine for games and other home uses, it is more a
>> marketting change than anything else - MS could easily have marketted W2K
>> for home use.
>
>W2K needs an upgrade as does any OS one year old. So, WinXP is an upgrade.
>MORE importantly, MS no longer will need to split it's resources between
the
>old and the new. AND even more importantly all the software writers out
>there will rejoice. There is only one version of the Windows API to program
>for. There is only need for one version of any given application. This is a
>good thing for everyone.
>
It is a good thing - I am not arguing that. I just can't see why you are so
impressed that MS are finally getting the finger out and fixing a problem of
their own creation that has been an enormous burden to everyone else since
Win95 first arrived. I also can't see why you, despite all the historical
evidence against it, are so ready to believe MS this time.
>>
>> >XP is a godsend for tech support. No longer having to ask: what version
>of
>> >windows are you running? and then having to fork your knowledgebase and
>> >script based on that.
>>
>> What's the weather like on your planet? Perhaps companies with one
>computer
>> will upgrade all at once, but for everyone else, there will be a mixture
>of
>> versions as always. Newer versions come in at the top, older ones go out
>at
>> the bottom, but the release of XP does not change anything there. It may
>> mean that eventually tech support will only have to ask which version of
>XP
>> people are running.
>
>Obviously not everyone will upgrade at once but this feature alone is one
>that might propel those that have been waiting even to do a W2K upgrade to
>consider going right to XP. Remote desktop take-over is even more
>compelling.
The problems, costs, and reluctance to upgrade to W2K are more likely to
make people even more hesitant to upgrade to XP than they were in upgrading
to W2K. Although in many cases W2K works fine, there are still too many
ways to trash it completly (especially using old hardware or software).
People looking for solid systems are still looking at NT 4 - it has its
problems, but they are better known. W2K is still considered by many to be
too immature for use as a business OS. W2K has been out for a year - people
are waiting for W2K.1, not a new way to play games on a business PC with a
GUI that is childish enough to make an eight-year-old cringe.
>
Do you know why the split between the two Windows families has continued so
long? There is no technical reason for it (as I mentioned above) - it is
purely marketting and economics. The home market will not pay as much as
the business market, so there had to be two different products to ensure
that each pay as much as possible. Eventually, it is getting to the stage
where they cannot keep up the pretence any longer, and are trying to find a
new way to differentiate the products (what was it? XP Personal and XP
Professional - one code base, two price tags), until they can get their
software renting system in action properly.
>
>>
>> Don't forget that XP is not yet ready - the "old crap" is MS's current
OS.
>
>True (old crap <= WinME) - but it's coming...
>
>
>
------------------------------
From: Dave Berntson <[EMAIL PROTECTED]>
Crossposted-To: alt.linux,comp.os.linux.networking,comp.os.linux.setup
Subject: Re: Drive Partitions
Date: Mon, 07 May 2001 12:47:42 -0500
On Mon, 7 May 2001 00:57:29 -0700, "A. Smith" <[EMAIL PROTECTED]>
wrote:
>
>I am trying to install Mandrake on a booty (lame) computer, and I get a
>bizarre error after seting up hard drive partions:
>"Not enough swap space for install."
>What does that mean? Does anybody know?
>The machine only has 16 MG or Ram, is that why?
>Any help would be apreciated.
>
>Andrew Smith [EMAIL PROTECTED]
>Software Developer http://catalyst.washington.edu
>Educational Technology Group http://depts.washington.edu/edtec
>University of Washington http://www.washington.edu
I had a similar error message during RH 7.0 install, but the error message
included an announcement that it was going to enable my swap partition to
complete the install.
If you are dual booting with windows, you might want to use fips to split
your hard drive before starting linux installation. Then use Disk Druid
during installation to create a swap partition (along with a root and a boot
partition, at a minimum). I had 32mb of ram, and in each of two
installations, I got some kind of error message and announcement about not
enough memory and enabling the swap partition.
My swap partition (created by Disk Druid during installation of linux) was
about 40 mbytes, IIRC.
Dave
------------------------------
From: "David Brown" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.ms-windows.nt.advocacy
Subject: Re: bank switches from using NT 4
Date: Mon, 7 May 2001 20:43:35 +0200
Jan Johanson wrote in message <3af69816$0$12207
>tiny itch of competition from them when usually the Japs take an american
>idea, improve and repackage it and beat us at our own game.
>
As compared to the Americans, who take a European idea, step up quantity and
better marketing while dropping quality down the toilet, and spend so much
on marketting that people think they invented it in the first place. Just
think of television, computers, hamburgers, cola, organized crime, ...
To be fair, there are a good number of ideas and inventions that have come
out of American, but I've got news for you - there are lot of ideas and
inventions that came from Japan too.
------------------------------
From: "David Brown" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.ms-windows.nt.advocacy
Subject: Re: bank switches from using NT 4
Date: Mon, 7 May 2001 20:50:37 +0200
T. Max Devlin wrote in message ...
>Said Jon Johansan in comp.os.linux.advocacy on 3 May 2001 11:49:14
> [...]
>>Imagine this:
>>XP is the achievement of a single code base. One set of drivers, you do
not
>>need to maintaine different drivers for W9x and ME and W2K. There is only
>>one set of updates. Only one GUI to learn. One way to do things. It's the
>>termination of a KNOWN ugly line of code. It's the end of ANYTHING
remotely
>>to do with DOS (other than emulation for backwards compatibility).
>
>Earth to Jon: that's what they said about Win98 and W2K, too. (And lots
>of it was said about Win95 and NT, as well.)
Not quite - Win95 was to be the last in the DOS+Win3.1 line. So was Win95
OSR2, Win98, Win98SE, and Windmill - each has been touted as the last of its
line. NT 4.0 was supposed to be for most users, but not quite all - it was
not claimed to be *entirely* backwards compatible. But W2K (and the NT5.0
that was promised as soon as NT4 arrived, but didn't make it until W2K) was
definitely claimed to be the uniting version of Windows. The trouble was,
it kept getting delayed and they needed the upgrade revenues - hence
Windmill was released. It will be interesting to see what happens if MS
fails to release a useable XP this year - they may end up having to release
a WindmillSE to keep up the profits until XP is ready.
>
>>XP is a godsend for tech support. No longer having to ask: what version of
>>windows are you running? and then having to fork your knowledgebase and
>>script based on that.
>
>So if everybody upgrades to the newest version, Microsoft promises that
>will finally be the end of the upgrade merry-go-round? Guffaw.
>
>>I don't see XP as a wait for it or think about it upgrade, I see XP as a
>>must have upgrade. Give me a shop running W2K servers and W2K/XP desktops
>>and I'll show you one that has cut tech support by half just from
>>eliminating support for old crap.
>
>Sock puppets. Geez.
>
>--
>T. Max Devlin
> *** The best way to convince another is
> to state your case moderately and
> accurately. - Benjamin Franklin ***
------------------------------
From: "Erik Funkenbusch" <[EMAIL PROTECTED]>
Subject: Re: How to hack with a crash, another Microsoft "feature"
Date: Mon, 7 May 2001 13:56:28 -0500
"Chad Everett" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> On Mon, 7 May 2001 05:11:24 -0500, Erik Funkenbusch <[EMAIL PROTECTED]>
wrote:
> >
> >You're forgetting. I already offered that it would be quite possible to
> >device an encoding scheme where the same sequence of characters are not
> >encoded the same way twice. Thus, all your occurances of A and O would
be
> >different for each time they occured. Suppose I used the 4 bit key as an
> >index into a completely arbitrarily chosen letter translation table.
>
> How do you distribute the translation table?
You don't need to. Both sides have the table. We're not talking about PKI
here, we're talking about two sides (say the Pentagon and some foreign
embassy). You can have totally unique algorithms for each embassy, so that
should one become compromised, your other embassies aren't. It's only
purpose is encryption between point a and point b, never point c, d, or e,
etc...
> > [ - snip - ]
> >
> >Further, the message could be written in an imaginary language (or one
known
> >only to the author and recipient, such as Apache which was used during
the
> >WWII). That in itself is a form of encoding. You might be able to
assume
> >that the imaginary language uses typical human language patterns, but if
the
> >language were made specifically to defy all known language patterns, you
are
> >again lost.
> >
>
> It was Navaho, not Apache.
I stand corrected.
------------------------------
Crossposted-To: soc.men,soc.singles,alt.fan.rush-limbaugh
From: <[EMAIL PROTECTED]>
Subject: Re: Why Linux Is no threat to Windows domination of the desktop
Date: Mon, 7 May 2001 18:56:49 GMT
>>>>> Aaron R Kulkis writes:
Aaron> [EMAIL PROTECTED] wrote:
>>
>> >>>>> Roberto Alsina writes:
>>
>> >> [EMAIL PROTECTED]> wrote:
>> >>>>>>> Aaron R Kulkis writes:
>> >>
Aaron> chrisv wrote:
>> >> >>
>> >> >> "Aaron R. Kulkis" <[EMAIL PROTECTED]> wrote:
>> >> >>
>> >> >> >> :The vast majority of human beings are bisexual, with most of them
>> >> >> >> :having a preference for one sex or the other.
>> >> >> >
>> >> >> >If your claim were true, then most bars, at the end of the night,
>> >> >> >would be the majority of men who failed to snag a women to hook
>> >> >> >up with each other and go home.
>> >> >>
>> >> >> No, that's not true at all.
>> >> >>
>> >>
Aaron> Really? If the majority of men are bisexual, then why aren't they
Aaron> hooking up with other guys after failing with the women?
>> >>
>> >> How many men in jail for long terms voluntarily engage in same
>> >> sex sexual acts? That percentage might give you an idea of
>> >> how many are to some degree bi-sexual.
>>
Roberto> Trading sex for protection != being bisexual
>>
>> Hence my choice of the word "voluntarily"
>>
Roberto> Being raped != being bisexual
>>
>> Hence my choice of the word "voluntarily"
>>
Roberto> Being desperate because you will never again see a woman != being bisexual
>>
Roberto> What you propose is like trying to determine if men like killing wild pigs
Roberto> based on episodes of "survivor".
>>
>> Nope, hence my choice of the word "voluntarily"
Aaron> And when you refer coercion and rape, the amount of VOLOUNTARY
Aaron> bi-sexuality among prison populations is....exceedingly low.
Cite?
--
Andrew Hall
(Now reading Usenet in alt.fan.rush-limbaugh...)
------------------------------
From: [EMAIL PROTECTED] (Bob Hauck)
Crossposted-To:
comp.os.ms-windows.nt.advocacy,comp.sys.mac.advocacy,comp.os.ms-windows.advocacy
Subject: Re: Justice Department LOVES Microsoft!
Reply-To: hauck[at]codem{dot}com
Date: Mon, 07 May 2001 19:04:25 GMT
On Mon, 07 May 2001 16:41:13 GMT, Daniel Johnson
<[EMAIL PROTECTED]> wrote:
> "Bob Hauck" <[EMAIL PROTECTED]> wrote in message
> news:[EMAIL PROTECTED]...
> > GDI calls->printer driver->printer
> >
> > And in Linux:
> >
> > Postscript->ghostscript->printer
> This isn't accurate. Information flows *both ways*
> for GDI.
Yes, and I mentioned that as an advantage of GDI down below.
> I expect the GS "driver" is really a user mode
> program that embeds a bitmap into appropriate
> printer language.
Yes, that's basically it. I don't know for sure what it does with
things like PCL5. It could be (and I think it is) more sophisticated
than a bitmap. In any case, I do know that my PCL printers work fine
with it. The output is not distinguishable from that produced by GDI
from the same input (e.g. by PDF documents).
> It's part of the basic problem with GS: GS information flows
> from the app to the printer, only. Some features just can't
> be supported like that.
I think I said that. I just didn't happen to think of the specific one
you came up with.
> The problem is more general than that, though. Consider
> printing on your HP PCL printer.
> But, as with PS, the printer has its own fonts. These
> fonts vary from printer to printer- some HP printers
> let you install more fonts using cute little cartidges.
Those same printers also let you install Postscript the same way.
That's a much better use of the slot, I think. Font cartridges are a
not-particularly-good idea that was popular for a while because of
software limitations.
> This means that GS has to render everything itself,
> using fonts that at least have the right metrics (does
> it use genuine Adobe fonts?); it can't just translate
> to PCL.
PCL and Postscript both support the concept of downloadable fonts. I'm
pretty sure GS will take advantage of that if it is available. Even if
it does not, it _could_ and so this is not a fundamental limitation.
> GDI gives the application access to printer specific
> font metrics; you can use them not only when printing
> but when laying out text for screen display. This gives
> you a shot at WYSIWYG even when font substition
> happens.
Ok, that's an actual advantage. But it also has disadvantages in that
you a) probably don't have exactly matching fonts for the screen, so
your layout may not be what you expect; and b) have now created a
document that nobody else can print properly unless they have the same
printer + cartridges as you do.
It is not clear to me that this is better than downloading the fonts or
printing a bitmap.
> The PS approach also limits you to printers
> that *can* emit PS or bitmaps; if you've got some
> lousy printer that just outputs text, GDI
> can handle it (forget WYSIWYG though) and
> GS can't.
Sure it can. The script "ps2ascii" uses GS to extract the ascii text
from a ps file.
--
-| Bob Hauck
-| Codem Systems, Inc.
-| http://www.codem.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
******************************
