Linux-Advocacy Digest #982, Volume #26 Thu, 8 Jun 00 19:13:06 EDT
Contents:
Re: Segmentation Fault? (R.E.Ballard ( Rex Ballard ))
Re: What else is hidden in MS code??? ("Drestin Black")
Re: What else is hidden in MS code??? ("Drestin Black")
Re: Some advocacy. (Steve Mentzer)
Re: The Mainframe VS the PC. (David E. Thomas)
Re: MacOS X: under the hood... (=?ISO-8859-1?Q?Lars_Tr=E4ger?=)
Re: There is only one innovation that matters... (=?ISO-8859-1?Q?Lars_Tr=E4ger?=)
Re: There is only one innovation that matters... (=?ISO-8859-1?Q?Lars_Tr=E4ger?=)
Re: OpenBSD security Re: Microsoft migrates Hotmail to W2K (Timothy J. Lee)
Re: The Mainframe VS the PC. (JEDIDIAH)
Re: The Mainframe VS the PC. ("Christopher Smith")
Re: The Mainframe VS the PC. (Steve Mentzer)
----------------------------------------------------------------------------
From: R.E.Ballard ( Rex Ballard ) <[EMAIL PROTECTED]>
Subject: Re: Segmentation Fault?
Date: Thu, 08 Jun 2000 22:04:45 GMT
In article <[EMAIL PROTECTED]>,
"1$Worth" <"1$Worth"@costreduction.removeplse.screaming.net> wrote:
> [EMAIL PROTECTED] wrote:
> >
> > You are running Linux.
> > Seg faults are normal operation for Linux.
Put more accurately, segmentation faults are managed and cause very
little significant hardship in the Linux environment.
> > See the SegFault How-To for information
>
> SegFaults should not be normal!
In the ideal world you are correct. In the real world of thousands
of frequently changing applications, these problems do pop up. How
you handle it is the issue.
Linux (and UNIX) will give you the option of creating a core dump
You can combine this with the source code to determine the cause,
resolve the problem, and create a source code change that can be
included in the next monthly freeze release which will be included
in the next quarterly distribution release.
Windows 3.1 would give you a "General Protection Fault" and you
knew it was time to save what you could and reboot your machine.
OS/2 2.0 would give you a Trap 0E or Trap 0D and you'd rebuild the
system.
Windows NT 3.51 would simply lock up with help from lots of nice
ugly messages.
OS/2 Warp 3.0 would trap, but would save enough that you could rescue
the system and regenerate or restore a saved desktop.
Windows 95 would just drop out, usually blow its guts, and lock up
waiting for you to reboot.
Windows NT 4.0 gave us the Blue Screen of Death. What made NT 4.0 fun,
especially as a server is that when you had an exception fault, it
would take down hundreds of concurrent users at the same time.
With Windows 2000 fabrics, you can mitigate the damages quite a bit.
The thread will terminate, but the rest of the application may still
be functional. You'll find out real quickly if not.
> Camilo, this signifies that either
> a programmer has screwed up their use
> of pointers or a physical memory problem
> or a hardware problem.
The one time I had frequent segfaults, it was because the cooling
fan on my Pentium had gotten blocked up. I moved the wire, made
sure the fan was working properly, sealed the box, and everything's
good on Linux.
> If you keep getting unexpected errors
> then have a look at the sig11 (segfault)
> faq at:
> http://www.BitWizard.nl/sig11/
>
--
Rex Ballard - Open Source Advocate, Internet
I/T Architect, MIS Director
http://www.open4success.com
Linux - 90 million satisfied users worldwide
and growing at over 5%/month!
Sent via Deja.com http://www.deja.com/
Before you buy.
------------------------------
From: "Drestin Black" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.ms-windows.nt.advocacy
Subject: Re: What else is hidden in MS code???
Date: Thu, 8 Jun 2000 18:11:35 -0400
"Rob S. Wolfram" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> Cheers,
> Rob
> P.S., should I congratulate you or express my condolences on
> http://usvms.gpo.gov/ms-final2.html ? ;-)
> --
um... why? do you suggest it's over?
this judge has been overturned on every single MS ruling he's made ever. Why
would this change now? hahaha...
------------------------------
From: "Drestin Black" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.ms-windows.nt.advocacy
Subject: Re: What else is hidden in MS code???
Date: Thu, 8 Jun 2000 18:15:40 -0400
"Rob S. Wolfram" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> [added cola so they can join the fun]
> Drestin Black <[EMAIL PROTECTED]> wrote:
> >"Rob S. Wolfram" <[EMAIL PROTECTED]> wrote in message
> >news:[EMAIL PROTECTED]...
> [... two months ago ...]
> >> There's enough independent peer review of the Linux source code by many
> >> independent individuals to *GUARANTEE* that there is no backdoor in
> >> Linux and all it's commonly used open source applications. There is no
> >> way you will ever be able to give an equal guarantee for any closed
> >> source application.
> >
> >Say Rob - how can I make a claim against that "*GUARANTEE*"? There is a
nice
> >big vulnerability in Open Source PGP (a commonly used open sores(tm)
> >application) in the linux version (not the windows version).
> >
> >1 year open source and all those peer reviews never spotted it?
>
> Sorry for the late reply, it's been a while since I followed comna. But
> there are a few remarks I want to make here.
>
> First, how is this a back door? I never made any claim about bugs,
> security wise or not. I made a claim about back doors. So let me
> challenge you. I can sell you the official version of Debian GNU/Linux
> (stable a.k.a. slink, this is 2 binary CD's and 2 source CD's and
> constitute *only* DFSG free software) for the amount of US$5000.00 This
> contains 2200 packages of free software. If you can find even *one*
> backdoor in any of those packages, I'll return you triple the amount.
> Oh, and selling you this software without support or manuals for such an
> amount is perfectly legal. Wanna play?
>
oooooo, lesse, so, for $5000 you'll give me all this stuff and invite me to
find any backdoor in any of these 2200 packages. One single backdoor in any
of these and you'll pay me $15,000?
if this is right, I would like to have my attorney write this up and send it
to you. OK?
Oh, and we'll do it this way: I'll put $5000 in an escrow account and YOU
will put $15,000 in that same escrow account (drawing interest in your
favor, of course, you can even keep the interest afterwards). Just so we
know we're all serious here. All we need to do is find any thing that is a
backdoor (and since you claim the FP97 dll thingy is a backdoor, I just need
to look for someplace where someone wrote something similar to "netscape
engineers are weenies" backwards and that will qualify) and I get the
$15,000 plus my original $5,000. Oh, and the results will be published here.
Wanna play?
------------------------------
From: Steve Mentzer <[EMAIL PROTECTED]>
Subject: Re: Some advocacy.
Date: Thu, 08 Jun 2000 15:19:42 -0700
>I cannot imaging programming using any sort of a GUI system. I can
>undeerstand how you might get flexibility, but only at the expense of
>being cumbersome. If you can give a short explanation of how such
>things help you write and compile code, please do. I write research
>physics code using F90 whose job is number crunching (3D turbulence,
>plasma physics, more or less fluid equations plus electrodynamics).
>
>Graphics with fancy systems I can see, programming I cannot.
You and I obviously live in different worlds in terms of programming
requirements. The vast majority of my programming work revolves around
client-server database applications and web presentation.
MS makes the architecture, implementation and deployment of these
applications quite easy. If it is done well, they can be very
reliable. It is purely dependent on the quality of the programmer.
>
>>Case in point: I can write a data-access component in 30 minutes that
>>will connect to *any* OLE-DB or ODBC compliant database. I can publish
>>this component on a server (via MTS/Component Services), apply Win2k
>>integrated security to the component, and generate a "point and click"
>>install script.
>
>Is this real programming or just module assembly?
Oh, this is real programming. It may not be the same a the modeling of
plasma physics, but it is a critical business need none the less...
>
>>Everything I just mentioned can be accomplished in Unix, but nowhere
>>near as easy or integrated.
>
>Why is "integrated" valuable? I program and write using about 8
>separate applications, each of which I jump back and forth via their
>respective windows. Fast, flexible, reliable.
>
It is a great feature to be able to distribute a standard component or
server that integrates with any OLE compliant windows application. In
most cases, if your data is persistable to a standard format, *any*
windows application that supports drag/drop cut/paste can use it.
>
>>I find that when using IIS, the *quality* of programming staff has a
>>lot to do with the effectiveness, efficiency and reliability of the
>>entire application.
>
>Don't even know what these things (IIS) are.
>
Internet Information Server. It is a middle-ware platform for hosting
various web services.
>>How so? NT has been rock solid for years, provided that you don't use
>>buggy drivers, and follow some basic rules when developing
>>applications.
>
>I hear wildly varying reports about NT. I am happy with what Solaris
>has become.
Same with win2k. It has been very stable since release.
>
>>The "ILOVEYOU" virus preyed on the compu-stupidity of the general
>>populous. If there was a unix email client that supported the in-line
>>execution of shell scripts, unix would also be succeptible to these
>>issues.
>
>As long as Outlook Express is default and auto-exec of attachments is
>also default, you will never escape this. In the past, Linux and Unix
>mail programs have also had trouble with this (eg, open ports,
>sendmail), but the open nature of the community uncovered the problems
>and newer versions are presented with the undesirable defaults
>corrected. Trouble is still caused by legacy software, of course, but
>that problem is universal.
I have a default installation of outlook express and it does not
auto-launch the attachment unless you explicitly open it.
>
>>MS made a tradeoff about the core OS. They understood that for people
>>to gain acceptance of an OS, it had to be easy to install and use
>>software.
>
>The install part is a major lie by the MS PR machine. Most people do
>not have to install MS OSes, and those who do very often have problems.
>This includes new devices. MS stuff is only what it is because
>everything comes with it, not because it is somehow easy, which it is
>not.
Installing MSOffice, Wordperfect, Visio, TurboTax etc etc etc is quite
easy, and very few people experience problems.
Contrary to popular believe, MS doesn't bunde everything with windows.
:)
>The necessity of becoming su is obvious if you want to take advantage of
>what you noted is a plus (not being able to corrupt the system as a
>normal user).
>
Agreed, but people have a hard enough time understanding "logins"
without without having to understand "super user". The security of the
OS is a plus, but the ease-of-use issue becomes a problem for John
Doe.
>On the sysadmin bit, I am the sysadmin on my Linux PC... such tools will
>always require people who can read and understand directions.
We are talking about the average user here. I have yet to see any
Linux/Unix based software installation documentation geared at the
"newbie".
>
>>They want to put in the CD, point click, install, and run the
>>software.
>
>If it works. If it does not (friends with mice or OS upgrades, etc),
>then you are stuck because there is no explanatory documentation.
If it doesn't work, and it is a large distribution, there are usually
tech support channels to work with. You may have to wait in a phone
queue, but in many cases that is 100 times as much support you would
get from a competing linux distro.
>
>My experience is that a Windows install on a personal desktop, assuming
>you have to do it, is harder and much less transparent that a similar
>Linux install.
While redhat and compteting distros have really improved the install
processes, they are still nowhere as easy as a windows install.
>
>>However, Win2k gives me a secure, stable and efficient application
>>development/deployment platform. Where Linux gives me brute power and
>>flexibility, Win2k gives me integration and "ease of development".
>
>I don't trust it. Even if it worked. And I don't like integration.
again, we live in two different worlds. In my world, application
interoperability is a PLUS. :)
>
>>There are trade-offs to both platforms, and which one you use depends
>>on the project at hand!
>
>As a computational physicist, I'll stick to Linux (and Unix).
And as a client-server application developer, I will stick to windows
:)
>
>This isn't really rocket science anymore... serious 2D and even
>simplified 3D turbulence codes with which you can learn all the basic
>physics are now well runnable on simple Linux laptops. When I get done
>writing it up this is going to be student stuff :-)
------------------------------
Subject: Re: The Mainframe VS the PC.
From: David E. Thomas <[EMAIL PROTECTED]>
Crossposted-To: comp.os.ms-windows.advocacy,comp.os.ms-windows.nt.advocacy
Date: Thu, 08 Jun 2000 15:18:37 -0700
In article <8hovhq$h9n$[EMAIL PROTECTED]>, [EMAIL PROTECTED]
wrote:
>
>Challenge: I'm building a network. 500 users. I already have
all the
>parts except the host, which can be a Win2k Server, an AS400,
or a VAX.
>Give me your best price, with supporting documentation to
include make &
>model, features, and performance statistics, on a Win2k Server
for my
>500 users. And someone else give me a price on an AS400 for the
same
>role. I can already get the stats, etc. from IBM; they have
nothing to
>hide.
>
You'd have to define exactly what 'service 500 users' means
before this would be meaningful at all. Are we talking about 3
shifts of workers (less than 200 at any one time then) that need
to logon and check e-mail a couple times a day, store a few
files and print once in a while? Are we talking about 500 people
constantly hitting a huge database for 8 straight hours a day?
Someting in-between? There's no such thing as a 'typical'
network.
I'm not disputing your claim necessarily. My (undocumented)
feeling is that if a 400 can do everything you want it to do it
will have the lowest TCO (not necessarily lowest initial
hardware / software costs, however). I would think HP-UX would
come in the highest in initial costs. (The last one I saw went
into a 300 user network and cost $150K).
If you plan on using figures from IBM, however, you must accept
figures from Microsoft for the W2K network and that is something
I suspect you wouldn't want to do. Independent figures all the
way around would be best.
David E. Thomas
* Sent from RemarQ http://www.remarq.com The Internet's Discussion Network *
The fastest and easiest way to search and participate in Usenet - Free!
------------------------------
From: [EMAIL PROTECTED] (=?ISO-8859-1?Q?Lars_Tr=E4ger?=)
Crossposted-To:
comp.os.ms-windows.nt.advocacy,comp.os.os2.advocacy,comp.sys.mac.advocacy,comp.sys.be.advocacy,comp.os.ms-windows.advocacy,comp.sys.amiga.advocacy
Subject: Re: MacOS X: under the hood...
Date: Thu, 8 Jun 2000 19:55:52 +0200
Greg Cox <[EMAIL PROTECTED]> wrote:
> Microsoft greatly helped IBM and Apple in the 80s by producing many
> popular products for their platforms. Both IBM and Apple thought it in
> their best interest to involve Microsoft in writing software for their
> products long before they were introduced (the IBM PC for IBM and the Mac
> for Apple).
IBM and Apple greatly helped MS in the 80s by giving them the first shot
at producing the OS / the first apps for their platforms - Heaven knows
why, they didn't have an OS and no apps to speak of. If this hadn't
happened, Microdoft would be an also-ran today.
Lars T.
------------------------------
From: [EMAIL PROTECTED] (=?ISO-8859-1?Q?Lars_Tr=E4ger?=)
Crossposted-To:
comp.os.ms-windows.nt.advocacy,comp.os.os2.advocacy,comp.sys.mac.advocacy,comp.sys.be.advocacy,comp.os.ms-windows.advocacy,comp.sys.amiga.advocacy
Subject: Re: There is only one innovation that matters...
Date: Thu, 8 Jun 2000 19:55:47 +0200
Joe Ragosta <[EMAIL PROTECTED]> wrote:
> In article <8h23om$e71$[EMAIL PROTECTED]>, "Stephen S. Edwards II"
> <[EMAIL PROTECTED]> wrote:
>
> > Also, remember that the idea of the PC was avalanched by Apple, to be
> > sure, but corporate use was limited. It was only after IBM, whose name
> > was equivalent to "computer" decided to manufacture and market a PC, did
> > the PC become widely accepted. However, they would have been dead in the
> > water if it were not for Microsoft's quick thinking, so in a way, yes,
> > Microsoft was the driving force that brought PCs to the masses.
>
> Nonsense. If IBM hadn't picked MS, they would have picked someone else.
> Even if it took a few months longer to release CPM, it wouldn't have had
> a material effect on today's computer business.
AFAIK, the story went something like this:
IBM decided to build a personal computer, but use an OS from somebody
else. Because Gates Mum had connections with someone at IBM, they first
asked Bill. Being honest for a change, he said MS didn't have an OS, but
infact sold CP/M from Digital Research with some products.
There are multiple stories what happened then, what it boils down to is:
they tried to make contact with Gary Kildall (big cheese at DR) but
didn't get to him in person (for whatever reasons), they wouldn't tell
Kildall's wife what they were here for if she didn't sign a multi-paged
NDA - which she refused, and they never tried again. Instead they again
went to Bill Gates.
He now said okay, got the rights to a CP/M clone for cheap and sold it
to IBM for much more - not without reserving the rights to sell the OS
on their own. The same people who wouldn't give Mrs. Kildall the time of
day without the NDA merryly said yes, thinking nobody would ever be able
to clone the engeneering masterpiece the PC was. The rest is history -
and MS is trying everything to rewrite it.
Lars T.
------------------------------
From: [EMAIL PROTECTED] (=?ISO-8859-1?Q?Lars_Tr=E4ger?=)
Crossposted-To:
comp.os.ms-windows.nt.advocacy,comp.os.os2.advocacy,comp.sys.mac.advocacy,comp.sys.be.advocacy,comp.os.unix.advocacy,comp.os.ms-windows.advocacy,comp.sys.amiga.advocacy
Subject: Re: There is only one innovation that matters...
Date: Thu, 8 Jun 2000 19:56:00 +0200
ŽOlafur Gunnlaugsson <[EMAIL PROTECTED]> wrote:
> "Stephen S. Edwards II" wrote:
>
> > Microsoft's main innovation is quite obvious: putting lots of computing
> > power into the hands of general consumers. Who else, besides, Commodore,
> > Apple, IBM, or Atari has even attempted this?
>
> Osbourne, Sinclair, (Intergalactic)Digital Research, Newbrain, Acorn, BBC,
> Sord, Epson, Rising Star, DRI (holland not that other one :), Dragon, Tandy
> Radio Shack, Tatung, A, Atlantis, Victor, Apricot, Sharp, Randofin, Tangerine,
> Amstrad, Kuma, Adobe, etc etc etc the list is endless
>
> Well you asked..........
>
> I am going to include Memotech and NeXT cause those computers looked so damn
> cool, and sported the correct colour for computers..
You should mention Sinclair again ;-) Somone at Sinclair (Sir Clive
himself?) once said something like "We started the designs with
different colors, but they usually ended black" (ignoring the ZX80 and
the keyboards for the ZX81 and Speccy).
Lars T.
------------------------------
From: [EMAIL PROTECTED] (Timothy J. Lee)
Crossposted-To: comp.os.ms-windows.nt.advocacy
Subject: Re: OpenBSD security Re: Microsoft migrates Hotmail to W2K
Date: 8 Jun 2000 22:27:35 GMT
Reply-To: see-signature-for-email-address---junk-not-welcome
[EMAIL PROTECTED] (Leslie Mikesell) writes:
|In article <8h9bnc$4ri$[EMAIL PROTECTED]>,
|Timothy J. Lee <see-signature-for-email-address---junk-not-welcome> wrote:
|
|>|>vulnerable to the six recently discovered security bugs). named
|>|>is not running in a default OpenBSD install;
|>|
|>|This is the problem I have with their claims of security in the
|>|default setup - with all the services turned off, the system
|>|is useless.
|>
|>But the system admin is expected to turn on just what s/he needs
|>to use the computer for the intended purpose. So the computer
|>isn't running imapd, named, "sendmail -bd", and all kinds of other
|
|But I do need those services and I need them most on the
|machine that is most likely to contain data worth stealing.
Does every computer you have exposed to the internet need to run
imapd, named, "sendmail -bd", and other such programs? I wouldn't
expect any non-mail server to need imapd or "sendmail -bd", for
example.
|>I've never seen an out of the box computer that was configured
|>just right for what I need it for. Either I have to turn lots of
|>stuff off, or turn things on, or both.
|
|Of course, but making claims about the system being secure because
|it isn't running services by default doesn't give me a warm
|fuzzy feeling about running it when you need the services.
In the real world, a lot of security problems (on both Linux
and Windows, as well as various other OSes) are the result of
default-on settings for services that are later found to have
security bugs. Rather than affecting only the 2%, 10%, or 50%
of people who need those services (and who are probably more likely
to be watching for security alerts on them so they can patch them
quickly), they affect 100% of those who use the OS (many of whom
don't realize that they are running those services, so they aren't
looking for security alerts).
|>|I didn't think the earlier version was
|>|happy running as a no-root user.
|>
|>-u, -g, and -t are features of BIND 8 (that OpenBSD backported
|>to 4.9.7). Oddly, Sun removed these options when bundling BIND
|>8.1.2 with Solaris 7.
|
|Why not just run BIND 8?
OpenBSD claims that BIND 8 code is messy enough that they don't
want to have to go through it for security bugs, so they are staying
with BIND 4.9.7, which they claim to have gone through. Since their
4.9.7 does have -u, -g, and -t (and these are enabled by default if
you turn named on), only the relatively small percentage of people
who need some other feature of BIND 8 need to go get it.
Of course, if you are running Solaris, you'll want to get BIND 8
from source to get the -u, -g, and -t options.
--
========================================================================
Timothy J. Lee timlee@
Unsolicited bulk or commercial email is not welcome. netcom.com
No warranty of any kind is provided with this message.
------------------------------
From: [EMAIL PROTECTED] (JEDIDIAH)
Crossposted-To: comp.os.ms-windows.advocacy,comp.os.ms-windows.nt.advocacy
Subject: Re: The Mainframe VS the PC.
Date: Thu, 08 Jun 2000 22:28:42 GMT
On Fri, 9 Jun 2000 08:09:20 +1000, Christopher Smith <[EMAIL PROTECTED]> wrote:
>
>"JEDIDIAH" <[EMAIL PROTECTED]> wrote in message
>news:[EMAIL PROTECTED]...
>> On Fri, 9 Jun 2000 07:02:12 +1000, Christopher Smith <[EMAIL PROTECTED]>
>wrote:
>> >
>> ><[EMAIL PROTECTED]> wrote in message
>news:8hovhq$h9n$[EMAIL PROTECTED]...
>> >> In article <[EMAIL PROTECTED]>,
>> >> "Drestin Black" <[EMAIL PROTECTED]> wrote:
>> >
>> >> I suspect a few M$ drones will trot out the tired old lie about
>> >> "productivity" with the GUI and how "intuitive" it is and that's
>> >> supposed to "improve" things. <yawn> Been there, heard that. Get a
>> >> frimping stopwatch, people, and quantify!!! Never mind the obvious lie
>> >> that a picture is somehow "intuitive"! Just pull up some actual
>numbers
>> >> as to how much does this GUI cost, on a transaction-by-transaction
>> >> basis, in Real Life!! DO IT!! I dare you. Back up your claim with
>> >> proof. You can plainly see for yourself that the GUI adds a burden to
>> >> the system! Show me how it lowers the TCO!
>> >
>> >So, how's your 80x25 text mode image editor coming along ?
>>
>> As long as you represent the image as something human readable,
>> it's quite possible to manipulate an image in a 80x25 character
>> cell interface actually...
>
>"Possible" was not the issue.
Depending on the situation, it can even be most convenient.
...and "possible" was indeed the issue for the comment in question.
Simple operations on xpm icons can be most conveniently done in emacs.
--
|||
/ | \
Need sane PPP docs? Try penguin.lvcm.com.
------------------------------
From: "Christopher Smith" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.ms-windows.advocacy,comp.os.ms-windows.nt.advocacy
Subject: Re: The Mainframe VS the PC.
Date: Fri, 9 Jun 2000 08:33:38 +1000
"JEDIDIAH" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> On Fri, 9 Jun 2000 08:09:20 +1000, Christopher Smith <[EMAIL PROTECTED]>
wrote:
> >
> >"JEDIDIAH" <[EMAIL PROTECTED]> wrote in message
> >news:[EMAIL PROTECTED]...
> >> On Fri, 9 Jun 2000 07:02:12 +1000, Christopher Smith <[EMAIL PROTECTED]>
> >wrote:
> >> >
> >> ><[EMAIL PROTECTED]> wrote in message
> >news:8hovhq$h9n$[EMAIL PROTECTED]...
> >> >> In article <[EMAIL PROTECTED]>,
> >> >> "Drestin Black" <[EMAIL PROTECTED]> wrote:
> >> >
> >> >> I suspect a few M$ drones will trot out the tired old lie about
> >> >> "productivity" with the GUI and how "intuitive" it is and that's
> >> >> supposed to "improve" things. <yawn> Been there, heard that. Get
a
> >> >> frimping stopwatch, people, and quantify!!! Never mind the obvious
lie
> >> >> that a picture is somehow "intuitive"! Just pull up some actual
> >numbers
> >> >> as to how much does this GUI cost, on a transaction-by-transaction
> >> >> basis, in Real Life!! DO IT!! I dare you. Back up your claim with
> >> >> proof. You can plainly see for yourself that the GUI adds a burden
to
> >> >> the system! Show me how it lowers the TCO!
> >> >
> >> >So, how's your 80x25 text mode image editor coming along ?
> >>
> >> As long as you represent the image as something human readable,
> >> it's quite possible to manipulate an image in a 80x25 character
> >> cell interface actually...
> >
> >"Possible" was not the issue.
>
> Depending on the situation, it can even be most convenient.
>
> ...and "possible" was indeed the issue for the comment in question.
No, it was not. The context of the discussion was TCO, not "it can be done,
eventually, and cost be damned".
> Simple operations on xpm icons can be most conveniently done in emacs.
But not complex digital image editing.
------------------------------
From: Steve Mentzer <[EMAIL PROTECTED]>
Crossposted-To: comp.os.ms-windows.advocacy,comp.os.ms-windows.nt.advocacy
Subject: Re: The Mainframe VS the PC.
Date: Thu, 08 Jun 2000 15:33:42 -0700
>I suspect a few M$ drones will trot out the tired old lie about
>"productivity" with the GUI and how "intuitive" it is and that's
>supposed to "improve" things. <yawn> Been there, heard that. Get a
>frimping stopwatch, people, and quantify!!! Never mind the obvious lie
>that a picture is somehow "intuitive"! Just pull up some actual numbers
>as to how much does this GUI cost, on a transaction-by-transaction
>basis, in Real Life!! DO IT!! I dare you. Back up your claim with
>proof. You can plainly see for yourself that the GUI adds a burden to
>the system! Show me how it lowers the TCO!
The benefits of a GUI based OS are largely dependent on the nature of
the applications.
Powerpoint presentations, complex documents, graphic design and most
other "visual" applications have benefited greatly from the GUI
innovation.
I still prefer to use the command line to perform system maintenance
and perform file operations. But I couldn't imaging typing a long
complex document using wordstart or applewriter!
I depends on what you are doing. Sometimes, the GUI gets in the way.
But that doesn't mean that it is bad for all applications :)
>
>> >
>> > You can by an AS 400 or HP 9000 of Dec VAX for under $100,000
>> > and it will service 500 people!
>> >
>> > You can't buy the servers under a W2K environment for that!
>>
>> Um, absolutely you can. And for less. And for more people.
>
>Um... Either you're lying, you're stealing, or you're pulling your
>"facts" straight out of YOUR backside.
>
>Challenge: I'm building a network. 500 users. I already have all the
>parts except the host, which can be a Win2k Server, an AS400, or a VAX.
>Give me your best price, with supporting documentation to include make &
>model, features, and performance statistics, on a Win2k Server for my
>500 users. And someone else give me a price on an AS400 for the same
>role. I can already get the stats, etc. from IBM; they have nothing to
>hide.
Performance is subjective. It depends on what you are willing to
accomplish.
A medium range dell server with raid 36GB raid 5, 1GB RAM and dual
733Mhz processors will run about 15k (give or take a thousand).
You can buy client licenses in bulk, but I don't have the figures.
We currently have a dual 600Mhz machine with 1GB ram and a RAID5
config running for 200 users. It is the primary file and print server,
performs nightly backups and also hosts an intranet.
The highest peak load we have seen during normal business hours was
around 15% (an extremely large print job).
I have complete confidence that the machine I detailed can handle 500
or more users under WIn2k.
------------------------------
** 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
******************************
