Linux-Advocacy Digest #508, Volume #34 Mon, 14 May 01 11:13:04 EDT
Contents:
Re: Justice Department LOVES Microsoft! ("Daniel Johnson")
Re: Good Tex Pdf Files was Re: Is StarOffice 5.2 "compatible" w/MS Office
97/2000? (Steve Bellenot)
Re: W2K/IIS proves itself over Linux/Tux ("Ayende Rahien")
Re: W2K/IIS proves itself over Linux/Tux ("Ayende Rahien")
Re: How to hack with a crash, another Microsoft "feature" (Bob Hauck)
Re: Win 9x is horrid (Craig Kelley)
Re: How to hack with a crash, another Microsoft "feature" (Bob Hauck)
Re: How to hack with a crash, another Microsoft "feature" (Roberto Alsina)
----------------------------------------------------------------------------
From: "Daniel Johnson" <[EMAIL PROTECTED]>
Crossposted-To:
comp.os.ms-windows.nt.advocacy,comp.sys.mac.advocacy,comp.os.ms-windows.advocacy
Subject: Re: Justice Department LOVES Microsoft!
Date: Mon, 14 May 2001 13:28:11 GMT
"Edward Rosten" <[EMAIL PROTECTED]> wrote in message
news:9do3s1$bg$[EMAIL PROTECTED]...
> >> If it isn't cross platform, than there's a limit as to how good it can
> >> be.
> >
> > If it *is* cross platform, then there's a limit as to how good it can
> > be.
> >
> > This is just as true a statement.
>
> Not really. IE can never be the best browser for me simply because it
> will never run on my system. If it was cross platform, then it could be.
That would put a damper on anyone's enthusiasm for the
thing, I think. :D
On the other hand, if IE did run on your system, it would be
impeded by its need to support other systems. For one thing,
in IE's case, it's structured as a collection of COM objects for
Window's benefit.
This isn't the right thing for Unix- it would be better as a
collection of filtering programs. Then you could use those
programs from a shell script, for one thing.
But it's can't be both; the two structures are too different.
Microsoft chose to favor their own platform and the result
is a rather weak Unix IE but a rather strong Windows one.
They could have had it the other way around, or of course
they could have made it weak on both.
But *somebody* has to lose out, inevitably.
That is why cross platforms are never the best
apps, unless they favor one platform and treat
the others as second class citizens. That's what
Microsoft does, but it means their offerings
can never take full advantage of any platform
but Microsoft's own.
I'm not saying that MS has any problem with
that, mind you. :D
------------------------------
From: [EMAIL PROTECTED] (Steve Bellenot)
Crossposted-To: comp.unix.advocacy,alt.solaris.x86,comp.unix.solaris
Subject: Re: Good Tex Pdf Files was Re: Is StarOffice 5.2 "compatible" w/MS Office
97/2000?
Date: 14 May 2001 13:35:02 GMT
In article <[EMAIL PROTECTED]>,
Bob Tennent <[EMAIL PROTECTED]> wrote:
>On Mon, 14 May 2001 13:31:04 +0100, Edward Rosten wrote:
> >>
> >> This is false. Dvips has the power to use outline versions of the cm
> >> fonts. For the TeX distribution we have installed all it takes is doing
> >> dvips -Pcms foo.dvi
> >
> >Mine doesn't do that :-(
> >
> >However, it is still not useful for PDF since PDF doesn't have outlined
> >versions of the cmr fonts avaliable.
There are publicly available outline versions of the cm font family and
the -Pcms options says to use them, at least with dvips version I have.
(I didn't install it, but I understand it is one of the standard
distributions for an older version of solaris (2.5.1).) I could have
this wrong, but I think the outline fonts
are pfb fonts since a find at the top of texmf tree shows
./fonts/type1/public/cm/cmr10.pfb
as well as the standard
./fonts/afm/public/cm/cmr10.afm
./fonts/pfm/public/cm/cmr10.pfm
./fonts/source/public/cm/cmr10.mf
./fonts/tfm/public/cm/cmr10.tfm
<and 16 zillion cmr10.pk>
My understanding is that the AMS (American Math Soc) donated the outlines.
>
>PDF is a format, not a program. There are many ways to produce PDFs with type 1
>cm fonts embedded. Here are some of the possiblities:
>
>pdf(la)tex
>dvi to pdf using dvipdfm
>dvi to ps to pdf using dvips and Adobe distiller
>dvi to ps to pdf using dvips and ghostscript (ps2pdf)
While true, dvips will, by default, put it pk fonts which are ugly. You
have to do something to get the outline versions.
>
>Check out
>
>http://www.cl.cam.ac.uk/~mgk25/publ-tips.html#type1
>http://www.yandy.com/download/pdf_from.pdf
>http://www.phy.hw.ac.uk/~peckham/programming/pdf/385e.htm
>http://tug.org/applications/pdftex/
>http://www.math.uakron.edu/~dpstory/latx2pdf.html
>http://www.babinszki.com/distiller/
>http://www.pdfzone.com/rich/distillersettings1.html
>http://www.pdfzone.com/rich/fonts1.html
>
>Bob T.
--
http://www.math.fsu.edu/~bellenot
bellenot <At/> math.fsu.edu
+1.850.644.7189 (4053fax)
------------------------------
From: "Ayende Rahien" <Don'[EMAIL PROTECTED]>
Crossposted-To: comp.os.ms-windows.nt.advocacy
Subject: Re: W2K/IIS proves itself over Linux/Tux
Date: Mon, 14 May 2001 05:18:32 +0200
"Matthew Gardiner" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> > > From what I have read, it is meant to be the silver bullet to solve
many
> > > of Windows problems, aka DLL Hell, and the messy Win32 API. I would
be
> > > interested in seeing how long it will take for .net apps to appear.
It
> > > took some vendors, such as Lotus, 7 years to move their apps from
Win16
> > > API to Win32 API. DOn't expect any miracles soon, it will be a cold
day
> > > in hell when I see a .net app on the shelves one year after the
release
> > > of .net
> >
> > Ahm, not that hard.
> > Office will be .NET.
> > As will be VS.
>
> .net is a whole new set of API's, how quickly do you think, say, Corel or
> Lotus will start utilising these new "features"?
How soon did it take to Java's applications to appear?
------------------------------
From: "Ayende Rahien" <Don'[EMAIL PROTECTED]>
Crossposted-To: comp.os.ms-windows.nt.advocacy
Subject: Re: W2K/IIS proves itself over Linux/Tux
Date: Mon, 14 May 2001 05:20:21 +0200
"Chris Ahlstrom" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> Chad's assuming a linear relationship between the number of
> customers and the number of vulnerabilities found. Pretty
> stupid reasoning. The linear relationship is actually
> between the number of vulnerabilities and the number of
> vulnerabilities found.
Not linear relationship, I'm afraid.
The relationship is more complex, and invovle both number of vulnerabilties
and number of users.
------------------------------
From: [EMAIL PROTECTED] (Bob Hauck)
Subject: Re: How to hack with a crash, another Microsoft "feature"
Reply-To: hauck[at]codem{dot}com
Date: Mon, 14 May 2001 14:48:20 GMT
On Mon, 14 May 2001 02:10:54 GMT, T. Max Devlin
<[EMAIL PROTECTED]> wrote:
> Said Bob Hauck in comp.os.linux.advocacy on Sun, 13 May 2001
17:43:51
> >On Sat, 12 May 2001 20:10:58 GMT, T. Max Devlin
<[EMAIL PROTECTED]> wrote:
> >> Said Bob Hauck in comp.os.linux.advocacy on Fri, 11 May 2001
14:37:28
> >> >On Fri, 11 May 2001 14:00:17 GMT, T. Max Devlin
> >> ><[EMAIL PROTECTED]> wrote:
> >> >> To wit, an algorithm is a recursive computational procedure with a
> >> >> finite number of steps. Translation tables need not apply;
> >> >
> >> >You can implement a translation table as a series of if-else tests,
> >> >thus turning it into a "computational procedure". Does that make a
> >> >translation table into an algorithm?
> >>
> >> No, quite specifically. If-else tests are conditional processing, not
> >> algorithms. An algorithm is a purely mathematical construct. Two plus
> >> two doesn't equal seven IF...
> >Many mathematical functions are non-continuous. They are defined as
> >one thing over one interval, and something else over another.
>
> These would be mathematical units, integers of time, are they not?
That would depend on whether we were discussing discrete or continuous
mathematics. I didn't specify, and the reason I didn't is that it does
not matter what the units are. For your sake I will restate anyway.
There are functions defined over the real numbers that are defined by
an "if-else" test. I gave an example of one (that is, a "purely
mathematical construct" that contains an if-else test).
> That isn't conditional on any thing but the assumption the world will
> still exist at the end of the for...next loop.
Parse error.
> > The unit step is one example that comes to mind. Sounds like an
> > if-else test to me (if time<0 then value=0 else value=1).
>
> So 2+2=7 IF what?
Nonsensical. Apparently you didn't understand what I was talking
about. I'll try again. The unit step function is a fundamental of
signal processing. It is *defined* as:
0 if time < 0
1 if time >= 0
The units of time can be either real numbers or integers, depending on
whether you're talking about the continuous version or the discrete
version of the function.
To recap, you asserted that an algorithm can't contain a "translation
table". I pointed out that it was trivial to remove the table and
replace it with an equivalent series of "if-else" tests. You then said
that such tests are "conditional processing" and not "algorithms" and
that algorithms had to be based on "purely mathematical functions".
So I gave an example of a purely mathematical function that is defined
in terms of if-else tests. I wanted to show that I could meet your
test of being "purely mathematical" and still sneak in some
"conditional processing". Now you seem to be saying that algorithms
can only be based on specific kinds of math functions, ones that do not
contain any conditionals, but it is hard to say for sure since you
appear to be trying to avoid saying anything concrete.
> >> An algorithm in software does not and cannot use translation tables;
> >> every value must be calculated, not looked up, or it is not algorithmic.
[snip using infinite series vs lookup tables to calculate the sine of an
angle]
> No, since you derived what you falsely called a "translation table"
> from... what? A translation of something? No, from mathematical
> calculations.
Ok, so if I "calculate" the table, then it can be part of an algorithm,
because it is not a translation table but a lookup table. But if I "do
something that isn't calculation" to make the table, then it can't be
part of an algorithm because it is a translation table. Is that a
correct interpretation of your argument?
> Thanks for your time. Hope it helps.
You actually had it right in the first place (when you said "an
algorithm is a recursive computational procedure with a finite number
of steps"). You should have left it at that. Instead you went off
track into this "can't have a table in an algorithm" bullshit because
you wanted Erik to be wrong. I find the lengths of illogic you are
willing to go to in order to defend an absurd position quite amusing.
So in that sense, yes, it does help.
--
-| Bob Hauck
-| Codem Systems, Inc.
-| http://www.codem.com/
------------------------------
From: Craig Kelley <[EMAIL PROTECTED]>
Subject: Re: Win 9x is horrid
Date: 14 May 2001 08:50:35 -0600
Donn Miller <[EMAIL PROTECTED]> writes:
> Craig Kelley wrote:
>
> > Contrast with Microsoft, who openly admits that you'll need to send
> > them intimate knowledge of your machine if you wish to "activate" your
> > products in the future.
>
> I don't think that Microsoft is all that "BAD" as people make them out
> to be. It's just a big monopolistic software giant, but they're
> probably not "BAD" 100% of the time. However, because of their size and
> influence over the software industry, even a tiny grain of "BAD" from MS
> does a lot of damage.
>
> Unix is just one operating system that did things right the first time.
> It is simply the best OS on the planet. Windows isn't "bad" all the
> time, but it just doesn't have the outstanding minimalist layer-by-layer
> design that unix OSes have. Even if MS weren't one iota bad, their OS
> still isn't as good as unix, because unix doesn't sufer from the DLL
> HELL that Windows does.
I like most Microsoft products. Office is great; Windows NT is much
better than their previous operating systems -- It's the management of
the company that irks me most.
Sure, there are some problems with NT, but there are problems with any
operating system. NT can be livable after a few hours of installing
missing software; but it still doesn't reach the ease-of-use level of
UNIX (IMHO). I don't even bother with it anymore as I use Linux
exclusivly now.
The problem Microsoft has is that they are an operating system company
in an era where the operating system is becomming a ubiquitous item.
In a few years people aren't going to care what OS is installed;
they'll all come with every basic tool you need (internet,
word processor, spreadsheet, audio/video tools, etc.). Microsoft's
offering will be just another face in the crowd -- the real money will
be made on hardware.
Microsoft doesn't build computer hardware; if they started to build it
they'd upset Dell, Compaq, etc. They'll eventually have to buy one of
the x86 "clone" makers in order to keep up (unless the X-Box is so
outstanding that they don't need to, I suppose). But they are in a
tight spot in the meantime, and hence all the hubub over licensing and
activation.
--
It won't be long before the CPU is a card in a slot on your ATX videoboard
Craig Kelley -- [EMAIL PROTECTED]
http://www.isu.edu/~kellcrai finger [EMAIL PROTECTED] for PGP block
------------------------------
From: [EMAIL PROTECTED] (Bob Hauck)
Subject: Re: How to hack with a crash, another Microsoft "feature"
Reply-To: hauck[at]codem{dot}com
Date: Mon, 14 May 2001 14:53:52 GMT
On 13 May 2001 23:34:36 -0500, Chad Everett
<[EMAIL PROTECTED]> wrote:
> >>A translation table is a form of lookup table. The DES uses lookup
> >>tables. Does this make it not an alogorithm? The Knuth-Morris-Pratt
> >>string search uses a lookup table too. In spite of that, Knuth thinks
> >>it is an algorithm. Is he wrong?
> Are the lookup tables known by everyone?
Some are fixed (and known), some are computed at runtime. Max asserted
in his other post that this makes a difference.
--
-| Bob Hauck
-| Codem Systems, Inc.
-| http://www.codem.com/
------------------------------
From: [EMAIL PROTECTED] (Roberto Alsina)
Subject: Re: How to hack with a crash, another Microsoft "feature"
Date: 14 May 2001 15:05:08 GMT
Reply-To: [EMAIL PROTECTED]
On Mon, 14 May 2001 14:48:20 GMT, Bob Hauck <[EMAIL PROTECTED]> wrote:
>On Mon, 14 May 2001 02:10:54 GMT, T. Max Devlin
><[EMAIL PROTECTED]> wrote:
>
>> Said Bob Hauck in comp.os.linux.advocacy on Sun, 13 May 2001
>17:43:51
>
>> >On Sat, 12 May 2001 20:10:58 GMT, T. Max Devlin
><[EMAIL PROTECTED]> wrote:
>
>> >> Said Bob Hauck in comp.os.linux.advocacy on Fri, 11 May 2001
>14:37:28
>
>> >> >On Fri, 11 May 2001 14:00:17 GMT, T. Max Devlin
>> >> ><[EMAIL PROTECTED]> wrote:
>
>> >> >> To wit, an algorithm is a recursive computational procedure with a
>> >> >> finite number of steps. Translation tables need not apply;
>> >> >
>> >> >You can implement a translation table as a series of if-else tests,
>> >> >thus turning it into a "computational procedure". Does that make a
>> >> >translation table into an algorithm?
>> >>
>> >> No, quite specifically. If-else tests are conditional processing, not
>> >> algorithms. An algorithm is a purely mathematical construct. Two plus
>> >> two doesn't equal seven IF...
>
>> >Many mathematical functions are non-continuous. They are defined as
>> >one thing over one interval, and something else over another.
>>
>> These would be mathematical units, integers of time, are they not?
>
>That would depend on whether we were discussing discrete or continuous
>mathematics. I didn't specify, and the reason I didn't is that it does
>not matter what the units are. For your sake I will restate anyway.
>There are functions defined over the real numbers that are defined by
>an "if-else" test. I gave an example of one (that is, a "purely
>mathematical construct" that contains an if-else test).
Just in case T. Max is really this ignorant of mathematics, here is a
simple example:
sign(x) is 1 if x>0, -1 if x<0 and undefined at 0 (there are alternative
definitions that provide a 0 or 1 value at 0, but this one is pretty
common.
sign is a pretty common, and very useful function.
For example, you can define the absolute value of a number as x*sign(x),
even if it's more common to define sign as abs(x)/x, with a special case
at 0.
>> That isn't conditional on any thing but the assumption the world will
>> still exist at the end of the for...next loop.
>
>Parse error.
>
>
>> > The unit step is one example that comes to mind. Sounds like an
>> > if-else test to me (if time<0 then value=0 else value=1).
>>
>> So 2+2=7 IF what?
2+2=7 *if* 7=4. Of course 7=4 is pretty hard to achieve, but as soon
as we, say, redefine "+" as an operation in a set different than the
integers, such a thing may happen.
I won't bother creating an algebraic structure for that in a usenet post,
but it is pretty easy.
A simpler example is, of course: 6+6=3 with addition defined as usual
in the mod(9) space (quick: mod(9) space defined as the numbers 0-8,
addition is as usual, but the result is the remainder of dividing
the usual result by 9).
>Nonsensical. Apparently you didn't understand what I was talking
>about. I'll try again. The unit step function is a fundamental of
>signal processing. It is *defined* as:
>
>0 if time < 0
>1 if time >= 0
>
>The units of time can be either real numbers or integers, depending on
>whether you're talking about the continuous version or the discrete
>version of the function.
>
>To recap, you asserted that an algorithm can't contain a "translation
>table". I pointed out that it was trivial to remove the table and
>replace it with an equivalent series of "if-else" tests. You then said
>that such tests are "conditional processing" and not "algorithms" and
>that algorithms had to be based on "purely mathematical functions".
>
>So I gave an example of a purely mathematical function that is defined
>in terms of if-else tests. I wanted to show that I could meet your
>test of being "purely mathematical" and still sneak in some
>"conditional processing". Now you seem to be saying that algorithms
>can only be based on specific kinds of math functions, ones that do not
>contain any conditionals, but it is hard to say for sure since you
>appear to be trying to avoid saying anything concrete.
>
>
>> >> An algorithm in software does not and cannot use translation tables;
>> >> every value must be calculated, not looked up, or it is not algorithmic.
>
>[snip using infinite series vs lookup tables to calculate the sine of an
> angle]
>
>> No, since you derived what you falsely called a "translation table"
>> from... what? A translation of something? No, from mathematical
>> calculations.
>
>Ok, so if I "calculate" the table, then it can be part of an algorithm,
>because it is not a translation table but a lookup table. But if I "do
>something that isn't calculation" to make the table, then it can't be
>part of an algorithm because it is a translation table. Is that a
>correct interpretation of your argument?
>
>
>> Thanks for your time. Hope it helps.
>
>You actually had it right in the first place (when you said "an
>algorithm is a recursive computational procedure with a finite number
>of steps"). You should have left it at that. Instead you went off
>track into this "can't have a table in an algorithm" bullshit because
>you wanted Erik to be wrong. I find the lengths of illogic you are
>willing to go to in order to defend an absurd position quite amusing.
His position is even more bizarre, if we consider a very simple algorithm,
used by everyone, that contains a (ok, very simple) translation table:
binary multiplication[1].
Binary multiplication consists of applying the following table to
two binary numbers in a specific way:
0 0 0
0 1 0
1 0 0
1 1 1
>So in that sense, yes, it does help.
[1] The above explanation of binary multiplication is intended for T. Max's
education, not as an insult to the reading audience ;-)
--
Roberto Alsina
------------------------------
** 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
******************************
