Linux-Advocacy Digest #948, Volume #25            Tue, 4 Apr 00 21:13:06 EDT

Contents:
  Re: benchmark for speed in linux / windows ([EMAIL PROTECTED])
  Re: benchmark for speed in linux / windows ([EMAIL PROTECTED])
  Re: Linux stocks soar in aftermarket trading ([EMAIL PROTECTED])
  Re: Why Linux on the desktop? (Christopher Browne)
  Re: Why Linux on the desktop? (Christopher Browne)
  Re: Introduction to Linux article for commentary (Christopher Browne)
  Re: benchmark for speed in linux / windows (Jim Richardson)
  Re: benchmark for speed in linux / windows ([EMAIL PROTECTED])
  Sorry Microsoft, Facts Mean More Than Money On The Net (was: benchmark for speed in 
linux / windows (Mark S. Bilk)
  Re: Sorry Microsoft, Facts Mean More Than Money On The Net (was: benchmark for speed 
in linux / windows ([EMAIL PROTECTED])
  Re: The Failure of Microsoft Propaganda -was- So where are the MS   ("Leonard F. 
Agius")

----------------------------------------------------------------------------

From: [EMAIL PROTECTED]
Subject: Re: benchmark for speed in linux / windows
Reply-To: [EMAIL PROTECTED]
Date: Wed, 05 Apr 2000 00:37:16 GMT


I said ./ not /

Big difference.....

Try again.......

Steve



On 5 Apr 2000 00:04:15 GMT, [EMAIL PROTECTED] (David
Steinberg) wrote:

>[EMAIL PROTECTED] wrote:
>: >>Try finding /etc/ppp/options using find from the root directory and
>: >>see how long it takes.
>
>: >    It takes less than a second on my box, even using find.
>
>: I find that extremely difficult to believe.
>
>That's because you're an idiot, Steve.
>
>At this point, I'm really wondering whether you've managed to install
>and use Linux yet, in spite of all your public attempts.  If you
>could use Linux, you'd see that find is not slow.
>
>Here are the results from two machines I had on hand:
>
>On my Athlon 600, with 4.5GB used (18.7GB total mounted) diskspace...
>                                                        Time elapsed:
>[dave@taro dave]$ find / -name options 2> /dev/null
>/etc/ppp/options                                    <-- 0.33 seconds
>/mnt/windows/windows/options
>/usr/doc/ppp-2.3.10/sample/options
>/usr/lib/tkX8.0.4/help/tk/widgets/options
>[dave@taro dave]$                                   <-- 1.26 seconds
>
>Just for fun, on my 486 DX2-66, with 544MB used (1.2GB total
>mounted) diskspace...
>[dave@mego dave]$ find / -name options 2> /dev/null
>/etc/ppp/options                                    <-- 2.24 seconds
>/usr/doc/ppp-2.3.7/sample/options
>/usr/lib/tkX8.0.4/help/tk/widgets/options
>[dave@mego dave]$                                   <-- 7.79 seconds
>
>Note that the total times for the execution of find (1.26 seconds and 7.79
>seconds) were obtained by using the time command.  The times to actually
>find the desired file (0.33 seconds and 2.24 seconds) were estimated with
>a stop watch.  Of course, these times are totally meaningless, since they
>depend on where the file happens to sit in the directory tree (I only
>included them to address your particular challenge to find 
>/etc/ppp/options).  In any case, I can find any file on my K7 in under a
>second and a half, and on my 486 in under 8 seconds, using find.  Which I
>would never do, because I could just use locate.  And, as was previously
>mentioned, I'm never going to need to search the whole thing from /,
>since I KNOW that a configuration file like that will be under /etc.
>
>You're full of it, as usual.


------------------------------

From: [EMAIL PROTECTED]
Subject: Re: benchmark for speed in linux / windows
Reply-To: [EMAIL PROTECTED]
Date: Wed, 05 Apr 2000 00:38:06 GMT



Nope the point of the discussion is speed not capability.

And for David that is ./ not /

Steve

On 5 Apr 2000 00:05:15 GMT, [EMAIL PROTECTED] (Bob Tennent)
wrote:

>On Tue, 04 Apr 2000 22:00:56 GMT, [EMAIL PROTECTED] wrote:
> >On Tue, 04 Apr 2000 21:19:31 GMT, [EMAIL PROTECTED] (JEDIDIAH)
> >wrote:
> >
> >>    No, the shill is just abusing the system to the benefit of 
> >>    his little agenda. The sort of find he's doing on Windows
> >>    should never be necessary on Unix.
> >
> >Incredible. With every message you go down the drain farther. So now
> >because Windows does something better than Linux,(which is usually the
> >case) it's not necessary under Linux.
>
>As I've pointed out in an earlier post that everyone seems to be ignoring,
>Linux find can do much more than Windows find.  So it's not the case
>that you are comparing the same things.  There is a program
>called treescan that is useful for the rare occasions that a complete
>scan *is* necessary and is much faster than find (though less flexible).
>That would be a fair comparison I would say.  I can't do the comparison
>myself because I don't have Windows.  
>
>Bob T.


------------------------------

From: [EMAIL PROTECTED]
Crossposted-To: comp.os.ms-windows.nt.advocacy,comp.os.os2.advocacy
Subject: Re: Linux stocks soar in aftermarket trading
Reply-To: [EMAIL PROTECTED]
Date: Wed, 05 Apr 2000 00:39:16 GMT

Amen.... It will peak and sink faster than the Titanic.

Steve


On Tue, 4 Apr 2000 20:10:54 -0400, "JOGIBA" <[EMAIL PROTECTED]>
wrote:

>Microsoft makes more money in one hour than all the Linux companies made in
>the last ten years. Linux will never be a mainstream desktop OS .


------------------------------

From: [EMAIL PROTECTED] (Christopher Browne)
Subject: Re: Why Linux on the desktop?
Reply-To: [EMAIL PROTECTED]
Date: Wed, 05 Apr 2000 00:42:37 GMT

Centuries ago, Nostradamus foresaw a time when Jim Dabell would say:
>Wrong again, they are markup languages.  Scripting languages are
>complete (turing complete?  I forget the full term).  The only
>difference between interpreted ("scripting" in your words) and compiled
>("programming" in your words) languages is the implementation, as you
>can have compiled Perl and interpreted C, etc.

The "much abused" term would be "Turing Machine equivalent."

The CS theorists [used to?] get excited about building TM variations
with different numbers of tapes, symbols, and states, in the quest to
build "minimal" computing machines capable of "doing everything."

The basic point to the non-theorist is that if a particular "computing
environment" provides:
  a) Some reasonable form of conditional branching, and
  b) Some reasonable form of storing values that are computed,
then it can be used to compute anything that can be computed by a
Turing Machine, and thus can be treated as being "equivalent" to any
other powerful computing device.

That makes no assumptions about *performance* or *convenience,* as a
particular "computing environment" may prove to be ridiculous to *use*
due to being either slow or ridiculously tedious to program.

The critical point is, however, that Perl is *essentially* as powerful
as C, and Common Lisp is *essentially* as powerful as Python, and this
can be extended arbitrarily to any choice of two programming
languages.

In contrast, HTML provides neither:
a) A form of conditional branching nor
b) Any way of storing computed values.

The fact that one can embed ECMAScript or PHP3 (or other languages')
code inside HTML does not change the fact that HTML is not, *in and of
itself,* a language that is "Turing Machine-equivalent."  HTML+PHP3
may be TM-equivalent, but that is *not* HTML.

HTML is not "TM-equivalent," and *THAT* is the reason why I would
argue that "knowing HTML" does not correspond to being "able to
program."  I consider "programming" to be primarily associated with
Turing Machines (e.g. - involving branching/data storage), which seems
to me to not be an outrageous constraint.

I don't expect all programmers to be familiar with the Lambda Calculus
(which could be an alternative definition!), but "condition" and "data
storage" seem to me to be reasonable minimum criteria for defining
"programming."

By those criteria, writing in any TM-equivalent language *does*
represent programming, whether it be a "compiled" language, as C, C++,
and Common Lisp almost always are, an "interpreted" language, as most
Schemes appear to be, or a "scripted" language, where programs are
represented as text, but may be treated at runtime as either
"compiled" or "interpreted" forms depending on the language.

Note that by the criteria I describe, someone that writes code using
either VBSCRIPT or Lotus Command Language (Lotus 123 "macro language")
*IS* programming.  They may not be C, but that doesn't make them not
programming... 
-- 
"As long as there are ill-defined goals, bizarre bugs, and unrealistic
schedules, there will be Real Programmers willing to jump in and Solve
The Problem, saving the documentation for later.  Long live FORTRAN!"
[EMAIL PROTECTED] <http://www.ntlug.org/~cbbrowne/lsf.html>

------------------------------

From: [EMAIL PROTECTED] (Christopher Browne)
Subject: Re: Why Linux on the desktop?
Reply-To: [EMAIL PROTECTED]
Date: Wed, 05 Apr 2000 00:42:43 GMT

Centuries ago, Nostradamus foresaw a time when John W. Stevens would say:
>Matthias Warkus wrote:
>> HTML is
>> not Turing complete. It hasn't got switches or looping constructs.
>
>Correct . . . but off topic.  The assertion I made (that got snipped),
>is:
>
>Every interaction with a computer is "programming".
>
>CREATING the HTML is an act of programming . . . unless you are trying
>to prove that Human beings are not Turing complete?

It's clear that when someone "programs" things that involve Loops and
Data Storage, that this is programming.

If, in contrast, someone is merely attaching HTML attributes to bits
of text, this can be indistinguishable from writing a document using a
word processor.
-- 
"In other words  -- and this is the rock solid  principle on which the
whole  of the Corporation's  Galaxy-wide success  is founded  -- their
fundamental design  flaws are  completely hidden by  their superficial
design flaws." -- HHGTG
[EMAIL PROTECTED] <http://www.hex.net/~cbbrowne/lsf.html>

------------------------------

From: [EMAIL PROTECTED] (Christopher Browne)
Crossposted-To: alt.os.linux
Subject: Re: Introduction to Linux article for commentary
Reply-To: [EMAIL PROTECTED]
Date: Wed, 05 Apr 2000 00:42:46 GMT

Centuries ago, Nostradamus foresaw a time when Tim Haynes would say:
>John Hasler <[EMAIL PROTECTED]> writes:
>> Tim writes:
>> > Sometimes I wonder what's so great about this capitalist thing after
>> > all. Or about communism.  Or any other political Thing.
>> 
>> Capitalism and communism are about economics.  Politics is about violence.
>
>LOL! Good one. OTOH C&C are also about 'mindsets' though. But I like the
>one about politics :)

Capitalism and communism are both about *propaganda* about political
economics.

Neither are actually being expressed by any particular political
system, instead, political systems take little bits of them, scatter
those little bits into their political economies, and then people
focus on other little bits in order to hold Flame Wars.
-- 
It is considered artful to append many messages on a subject, leaving
only the most inflammatory lines from each, and reply to all in one
swift blow.  The choice of lines to support your argument can make or
break your case.
-- from the Symbolics Guidelines for Sending Mail
[EMAIL PROTECTED] <http://www.ntlug.org/~cbbrowne/lsf.html>

------------------------------

From: [EMAIL PROTECTED] (Jim Richardson)
Subject: Re: benchmark for speed in linux / windows
Reply-To: [EMAIL PROTECTED]
Date: Wed, 05 Apr 2000 00:44:48 GMT

On Tue, 04 Apr 2000 14:54:48 GMT, 
 [EMAIL PROTECTED], in the persona of <[EMAIL PROTECTED]>,
 brought forth the following words...:

>How about using find under Linux and find under Windows and see what
>happens.
>
>Linux churns away for an eternity and Windows has the result in a
>couple of seconds.
>
>I'm not talking about FastFind either, just the normal find that comes
>with Windows.
>
>Windows wins by a large margin, searching a similar number of files.
>
>Steve
>
>

Use the right tool for the job, use locate.

For example. I searched for a file a few levels deep in my home dir

jim@raven:~ > time locate kjSet.py
/home/jim/downloads/kwp/kjSet.py
/home/jim/inwork/Zope-2.1.4-src/lib/python/Products/ZGadflyDA/gadfly/kjSet.py
/home/jim/inwork/Zope-2.1.4-src/lib/python/Products/ZGadflyDA/gadfly/kjSet.pyc

real    0m0.756s
user    0m0.740s
sys     0m0.020s


I'd forgotten about the .pyc file also :)

So Steve, pick the right tool for the job, and you'll be much happier.

-- 
Jim Richardson
        Anarchist, pagan and proud of it
WWW.eskimo.com/~warlock
        Linux, because life's too short for a buggy OS.


------------------------------

From: [EMAIL PROTECTED]
Subject: Re: benchmark for speed in linux / windows
Reply-To: [EMAIL PROTECTED]
Date: Wed, 05 Apr 2000 00:49:48 GMT

The question is why is Linux Find so damm slow?

 Is it the disk i/o subsystem? 

The OS itself?



 What is the reason?

Why is Windows so much faster at a brute force find a file name
search?



Steve


On Wed, 05 Apr 2000 00:44:48 GMT, [EMAIL PROTECTED] (Jim Richardson)
wrote:

>On Tue, 04 Apr 2000 14:54:48 GMT, 
> [EMAIL PROTECTED], in the persona of <[EMAIL PROTECTED]>,
> brought forth the following words...:
>
>>How about using find under Linux and find under Windows and see what
>>happens.
>>
>>Linux churns away for an eternity and Windows has the result in a
>>couple of seconds.
>>
>>I'm not talking about FastFind either, just the normal find that comes
>>with Windows.
>>
>>Windows wins by a large margin, searching a similar number of files.
>>
>>Steve
>>
>>
>
>Use the right tool for the job, use locate.
>
>For example. I searched for a file a few levels deep in my home dir
>
>jim@raven:~ > time locate kjSet.py
>/home/jim/downloads/kwp/kjSet.py
>/home/jim/inwork/Zope-2.1.4-src/lib/python/Products/ZGadflyDA/gadfly/kjSet.py
>/home/jim/inwork/Zope-2.1.4-src/lib/python/Products/ZGadflyDA/gadfly/kjSet.pyc
>
>real    0m0.756s
>user    0m0.740s
>sys     0m0.020s
>
>
>I'd forgotten about the .pyc file also :)
>
>So Steve, pick the right tool for the job, and you'll be much happier.


------------------------------

From: [EMAIL PROTECTED] (Mark S. Bilk)
Subject: Sorry Microsoft, Facts Mean More Than Money On The Net (was: benchmark for 
speed in linux / windows
Date: 5 Apr 2000 00:58:34 GMT

In article <YWqG4.10645$06.32384@wards>,
Robert Moir <[EMAIL PROTECTED]> wrote:
>"Tim Kelley" <[EMAIL PROTECTED]> wrote in message
>news:[EMAIL PROTECTED]...
>> [EMAIL PROTECTED] wrote:
>> >
>> > How about using find under Linux and find under Windows and see what
>> > happens.
>> >
>> > Linux churns away for an eternity and Windows has the result in a
>> > couple of seconds.
>> >
>> > I'm not talking about FastFind either, just the normal find that comes
>> > with Windows.
>> >
>> > Windows wins by a large margin, searching a similar number of files.
>>
>> I would wager that typically windows "find" is sifting through a
>> LOT LESS than linux is.  If you are just looking for a filname
>> and specify something like "/" or even "/usr" linux find is
>> probably looking through a GB of files, even more, perhaps >2GB
>> if you installed one of the modern wiz bang distros and installed
>> everything.
>
>So Linux is so bloated that it's search is slow despite being super
>efficent?

Any person who is even slightly familiar with Linux distri-
butions knows that the 2GB of files referred to by Tim Kelley
mostly comprise about *2,000 application programs* that are 
included at almost no cost with the Linux operating system.  

Robert Moir has been posting Microsoft propaganda to this 
newsgroup for months, so he certainly knows this.  

Therefore, his false characterization of those files as 
"bloat" is a knowing and intentional *lie* on behalf of 
Microsoft, which he hopes will be believed by people who are 
thinking about trying Linux, and will mislead them into not 
doing so.

Organizations like Microsoft are used to telling their lies
without opposition in the print and broadcast media that 
their immense wealth controls.  So it must be really annoying 
to them that on the Internet, and especially Usenet, their 
lies are totally and embarrassingly refuted with facts by 
people who only need enough money to pay $20 a month for a 
connection.

Maybe that's why Microsoft has decided to discontinue Usenet
access for all of the customers of its Internet service
provider division, MSN.com.

(Note: Whether Moir is or is not getting paid in some way by
Microsoft is irrelevant to the above.)



------------------------------

From: [EMAIL PROTECTED]
Subject: Re: Sorry Microsoft, Facts Mean More Than Money On The Net (was: benchmark 
for speed in linux / windows
Reply-To: [EMAIL PROTECTED]
Date: Wed, 05 Apr 2000 01:08:29 GMT

On 5 Apr 2000 00:58:34 GMT, [EMAIL PROTECTED] (Mark S. Bilk) wrote:

>In article <YWqG4.10645$06.32384@wards>,
>Robert Moir <[EMAIL PROTECTED]> wrote:
>>"Tim Kelley" <[EMAIL PROTECTED]> wrote in message
>>news:[EMAIL PROTECTED]...
>>> [EMAIL PROTECTED] wrote:
>>> >
>>> > How about using find under Linux and find under Windows and see what
>>> > happens.
>>> >
>>> > Linux churns away for an eternity and Windows has the result in a
>>> > couple of seconds.
>>> >
>>> > I'm not talking about FastFind either, just the normal find that comes
>>> > with Windows.
>>> >
>>> > Windows wins by a large margin, searching a similar number of files.
>>>
>>> I would wager that typically windows "find" is sifting through a
>>> LOT LESS than linux is.  If you are just looking for a filname
>>> and specify something like "/" or even "/usr" linux find is
>>> probably looking through a GB of files, even more, perhaps >2GB
>>> if you installed one of the modern wiz bang distros and installed
>>> everything.
>>
>>So Linux is so bloated that it's search is slow despite being super
>>efficent?
>
>Any person who is even slightly familiar with Linux distri-
>butions knows that the 2GB of files referred to by Tim Kelley
>mostly comprise about *2,000 application programs* that are 
>included at almost no cost with the Linux operating system.  

Who the hell cares ?
The discussion is about speed of find under Linux and Windows. The
content of the files is not an issue as we are searching for file
name, nothing else.


>>Robert Moir has been posting Microsoft propaganda to this 
>newsgroup for months, so he certainly knows this.  

See above.....

>Therefore, his false characterization of those files as 
>"bloat" is a knowing and intentional *lie* on behalf of 
>Microsoft, which he hopes will be believed by people who are 
>thinking about trying Linux, and will mislead them into not 
>doing so.


Linux speaks for itself. I gave away 2 Linux CD's today. The suckers
begged me for copies and I was more than happy to oblige.
Just one look at Linux and they'll be back to Windows in a flash.


>Organizations like Microsoft are used to telling their lies
>without opposition in the print and broadcast media that 
>their immense wealth controls.  So it must be really annoying 
>to them that on the Internet, and especially Usenet, their 
>lies are totally and embarrassingly refuted with facts by 
>people who only need enough money to pay $20 a month for a 
>connection.

Linux supporters and Facts are an oxymoron. 

>Maybe that's why Microsoft has decided to discontinue Usenet
>access for all of the customers of its Internet service
>provider division, MSN.com.

Does anyone actually use MSN?


>(Note: Whether Moir is or is not getting paid in some way by
>Microsoft is irrelevant to the above.)


I thought we were all paid shills?

Where do I collect my check? I could use a few more gadgets for my
studio.

Steve




------------------------------

From: "Leonard F. Agius" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.ms-windows.nt.advocacy,alt.destroy.microsoft
Subject: Re: The Failure of Microsoft Propaganda -was- So where are the MS  
Date: Wed, 05 Apr 2000 01:09:16 GMT

[EMAIL PROTECTED] wrote:

> On Tue, 04 Apr 2000 23:44:15 GMT, [EMAIL PROTECTED] (JEDIDIAH)
> wrote:
>
> >On Tue, 04 Apr 2000 21:57:04 GMT, Leonard F. Agius <[EMAIL PROTECTED]> 
>wrote:
> >>fmc wrote:
> >>
> >>> "Mark S. Bilk" <[EMAIL PROTECTED]> > In article
> >>> <[EMAIL PROTECTED]>,
> >>> >  <[EMAIL PROTECTED]> wrote:
> >>> >
> >>> > >Linux truely speaks for itself. For every geek that loves the control
> >>> > >there are 500 normal users that need to accomplish tasks that require
> >>> > >software that simply is not available under Linux. Or if it is
> >>> > >available, it is so crude and ugly looking it is not worth mentioning.
> >>> > >Or it's simply not compatible with what the rest of the free world is
> >>> > >running.
> >>> >
> >>> > The true situation is that applications fulfilling the
> >>> > requirements (with the exception of games) of most Windows
> >>> > users are *now* available under Linux, almost all of them
> >>> > at no cost.
> >>>
> >>> Most people have some requirements that go beyond the standard
> >>> WP/Spreadsheet/Browser.  I need a  financial app like Quicken or MS Money, a
> >>> tax preparation program like TurboTax, TaxCut, or TaxSaver, and project
> >>> management software like MS Project or CA-SuperProject.  These don't exist
> >>> for Linux.  I also can't manage my bank accounts online.  That requires
> >>> either Windows or Mac.
> >>>
> >>> For myself, I'll wait to try Linux again until solutions for my needs become
> >>> available.  It will be a long wait if I have to rely on the open source
> >>> community to provide them.
> >>>
> >>> fmc
> >>
> >>fmc hit the nail right on the head. There aren't the off-the-shelf solutions for
> >>Linux, or any of the other OS's, save Apple/Mac. I can't port my scanner's
> >>software to Linux or BEOS, I can't get as inexpensive AND well supported (notice
> >>I qualified it with both inexpensive AND well supported) an image editor as
> >>Paint Shop Pro. I can't get a fax/voice mail solution like Win Fax or Talkworks
> >
> >       Gimp satisfies that criterion quite nicely actually. As far as
> >       'porting' something, that's your burden. There are abstractions
> >       available for that sort of thing (SANE vs. TWAIN).
>
> Don't make me laugh. My $69.00 Canon scanner came with enough "free"
> software to blow the doors off anything Linux has, including Gimp.
> Not to mention it worked perfectly out of the box.
> The wizards did everything from configuring to prompting me through
> making my first scan.
>
> Worked like a charm right out of the box and no overpriced SCSI
> needed.
>
> Sane?
>
> Should be called insane...What a joke.....
>
> Linux misses the boat again. When will you people understand that
> setup.exe is your friend?
>
> In this case all I did was pop the CD in and away it went.
>
> Steve

LOL!!! You tell 'em, Steve. When the hell are these people going to realize that the
majority of people WANT the MS Monopoly...they want a COMMON STANDARD, not a bunch of
incompatible Unix variants!!!


--
Fight SPAM!!! Remove the _nospam from the above address to send e-mail.

The opinions expressed are my own.



------------------------------


** 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
******************************

Reply via email to