Linux-Misc Digest #735, Volume #21 Thu, 9 Sep 99 05:13:09 EDT
Contents:
Re: Amiga, QNX, Linux and Revolution (Joe Cosby)
X clients-- allowing??? (Brian Joseph Carmody)
Re: What is best HTML Editor for LINUX? (Jon Skeet)
Re: weird: /dev/hda missing during lilo install (Leonard Evens)
Re: Making new boot disk (Leonard Evens)
Re: NIC stops responding (Leonard Evens)
Re: Q? - best combo of linux distrib and apps for 3rd world (David M. Cook)
Re: weird: /dev/hda missing during lilo install (Simon Oosthoek)
Re: signal 11 (Leonard Evens)
rplay oddity (Peter Schaffter)
rplay oddity (Peter Schaffter)
how to find font used by an xterm? (brian@com)
----------------------------------------------------------------------------
From: [EMAIL PROTECTED] (Joe Cosby)
Crossposted-To: comp.os.qnx,comp.sys.amiga.misc
Subject: Re: Amiga, QNX, Linux and Revolution
Date: 9 Sep 1999 06:07:11 GMT
** To reply in e-mail, remove "howsyz." from address **
Guy Macon hunched over his computer, typing feverishly;
thunder crashed, Guy Macon laughed madly, then wrote:
> In article <7qvpr3$v0v$[EMAIL PROTECTED]>, [EMAIL PROTECTED]
>(Joe Cosby) wrote:
> >
> >** To reply in e-mail, remove "rokwak." from address **
> >
> >Guy Macon hunched over his computer, typing feverishly;
> >thunder crashed, Guy Macon laughed madly, then wrote:
> >> In article <7qukv8$dvr$[EMAIL PROTECTED]>,
>[EMAIL PROTECTED] (Joe Cosby) wrote:
> >>
> >> >If the basic concepts (what is a file, what is a program, GUI
> >> >elements, directory structure) were introduced first, I think
> >> >real computer literacy would rise.
> >>
> >> This is the wrong direction to go. Files, programs, and directories
> >> are a classic case of the human interface being a one to one
> >> representation of the internal structure of the program. Calling a
> >> directory a Folder doesn't change that. I can learn how to drive
> >> a car or to use the telephone network without knowing the internal
> >> details. Who says that the way CP-M and Unix organized information
> >> is the most understandable for humans?
> >>
> >
> >Not the particular directory structure, but what is a 'directory
> >structure' at all.
> >
> >Given a mass storage device, files will be stored in a series of
> >directories, each directory can have subdirectories, etc.
>
> You are desribing the internal organization used by the system
> software. There is no reason why the human interface has to match
> the internal organization, and many reasons why you might want to
> organize the human interface differently. Look at what we are doing
> right now. This discussion is presented to me as newsgroups which
> contain articles. the directory structure and whether all the posts
> are seperate files or entries in a relational database. Look at the
> WWW. You and I know that it is based on Unix style directories, but
> many users only see hyperlinks and pages on their screens.
>
> >I'm talking about beginners, here.
>
> Me too.
>
> >People really don't know that. People can run Windows/MS software
> >and not know that.
>
> Nor do they have to. There was a time when you needed to understand
> spark advance to operate an automobile. Now it is quite possible
> to operate one without ever learning how to open the hood or to add
> gasoline to the tank.
>
> >People don't know what is going on when they 'start a program'.
> >They don't know what happens when you 'save a program'.
>
> Nor should they. When I work on a word processing document I am
> faced with the internal implementation detail of fast storage that
> dies when power is removed and doesn't require names (RAM) and
> slower storage that survives a power failure and requires filenames.
> Why should the user need to know any of this?
>
> >The human interface -should- be a one to one representation of the
> >system internals, economy and elegance are -always- preferable.
> >
> >An automobile driver presses the gas to accelerate, releases the
> >gas to cease adding power, presses the brake to decelerate, turns
> >the wheel to direct the car. There is a one to one relation
> >between the interface and the major functions of the system.
>
> There is a a one to one relation between the interface and the
> functions that the user needs to do (stop, go, steer), but there
> is not a one to one relation between the interface and the internal
> implementation details. Manual shifting is a good example of a one
> to one relation between the interface and the internal implementation
> details, and manual shifting makes it harder to learn how to drive.
> An ideal car would be an automatic that a power user could switch
> into manual shifting mode. An ideal car would also not give you
> two seperate ways to shed speed with different interfaces (brakes
> and engine braking). An ideal car would have the brake pedal control
> the brakes and the compression braking (with a power user mode that
> gives full control).
>
Here, I agree with you 100%. When I design an interface, I ask myself first, what is
the
user going to be trying to do when he accesses this interface? Second, what is the
ideal absolute logical -minimum- input I am going to need from him to accomplish that?
So for instance, we're manufacturing parts to fill a large order. The production
supervisor might want to use the computer to find out how many parts have been made
against the current order, so he knows how many more need to be made. So as I see
it, that can be reduced to two interactions;
1. I want to check production vs. need
2. Select part to check
So if I end up with more than two interactions, as I see it I've failed.
That is not only how I design interfaces for others; frequently that's how I design
them
for myself.
I'm not saying that people should be required to understand all the inner workings of a
computer in order to operate one.
I am saying that there are basic concepts in how computers work which are universal
across platforms.
We are requiring people to interact with these machines, but we are not teaching them
the basic concepts necessary to interact with them effectively. A hundred, maybe two
hundred years ago most people learned little or no mathematics. They didn't need to.
Now, it would be difficult to survive without at least enough math to manage a bank
account, decide whether a purchase was practical, etc.
Forty-fifty years ago, most people didn't drive autos, so driver's ed wasn't taught in
school. Now most people drive, so it is.
Ten-fifteen years ago, you didn't need to be computer literate to be employable, so
basic computer literacy wasn't taught. And it still isn't.
You're talking about changing what it means to be computer-literate. I'm talking about
teaching basic, inevitable concepts of computer use.
Maybe, as you say, hierarchical file structures should be abandoned in the interface in
favor of something 'less technical'. Less of an outgrowth of the computer's internal
structure.
And, OK, maybe so. But first: what I'm saying is, right now, this is how computers
work
and -until- that changes, I think people who are going to depend on these machines
should be taught core concepts for their own sake, just as they are taught core
concepts
of math and driving.
Second, is it really realistic to say that hierarchical directory structure is a bad
idea, or
an idea to be 'gotten beyond'? Computers process large amounts of data very quickly.
File cabinets can be replaced by desktop computers. Storage warehouses can be
replaced by LAN servers. A general computer operator is either going to be handling
large amounts of data, or is not going to be getting as much usefulness out of the
computer as she can.
So I think a hierarchical structure, or something analogous, is just a good logical
choice
for keeping 'papers' sorted.
Is it really such a difficult concept? People learned to use filing cabinets and
desks for a
hundred years with no problem. The concepts aren't really all that different, so long
as
you work at it from 'the ground up'.
Cooper provides a wonderful example. 'Jane' loses all her files because someone
changes her default directory in Word.
What I'm saying is, if Jane had been taught a basic core concept like directory
structure
she wouldn't have batted an eye.
And from the interface design level, what could be done differently, so that Jane
wouldn't 'have to worry about' structure? MS provide two good solutions. She might
have clicked the 'File' menu, and checked her last four files from within Word. She
might have clicked 'Start'->'Documents' from the patented Windows Upside-Down Menu
(tm) and found her last doc that way.
But I think the problem with these two solutions is immediately obvious. They -add- to
the amount of knowledge necessary for Jane to handle her computer.
So the next step seems to be to say "well, they should be her -primary- interface".
But as I see it, teaching hierarchical directory structure -once- in high school,
where this
is applicable across a wide variety of needs, just makes more sense.
And this also leads into my next point;
Third: If we get away from interfaces which are based on a 'best representation' of
the
system's internal structure, then I think we end up with a lack of uniformity across
applications, which is even worse for the user.
> >Even something as basic as 'how is data stored' can go a very long
> >way towards making a computer operator, however dullwitted, a much
> >better operator. When I first ran a Wintel, I had been using an
> >Amiga for years. It took me no time to find programs, run them,
> >store files, retrieve files, etc. These things take some people
> >months, and many people never really understand it. The basics
> >are universal (plus or minus a GUI).
>
> Why should you have to find programs? Why should you be aware of
> the existance of a hard disk?
>
I know a large number of people who have filled up their hard disk, and not understood
why performance dogged (as Windoze juggled page files).
To me it's like saying "why do you need a gas gauge? This car has a -really big- gas
tank.
> You REALLY need to read THE INMATES ARE RUNNING THE ASYLUM. It's
> amazon.com's best selling software development book. It even describes
> what an apologist is, which fits you (and me, before I read the book!)
> very well. Look into it. There is a better way to do things.
>
Despite all this which is disagreeing as far as the need for basic
computer education, I don't want to seem like I'm flaming or
something.
Really, I've found this whole thread very interesting and the
ideas you and Cooper are talking about very important and
insightful.
Ironic that the whole thread began with that drooling cretin 'Buzz
Olin' and his crossposting
:^)
--
Joe Cosby
Amiga Fanatic
------------------------------
From: Brian Joseph Carmody <[EMAIL PROTECTED]>
Subject: X clients-- allowing???
Date: Thu, 9 Sep 1999 02:04:48 -0400
I have X serving my localhost, but i also want to be able to run X
sessions with other machines on the network (instead of telneting into
them). Where is the conf file where you add the hosts with are allowed to
connect to your machine???? (Red Hat 6.0) Thanks in advance
------------------------------
From: [EMAIL PROTECTED] (Jon Skeet)
Subject: Re: What is best HTML Editor for LINUX?
Date: Thu, 9 Sep 1999 08:10:58 +0100
[EMAIL PROTECTED] wrote:
> Do you suppose we'll ever see an html editor that allows you to preview
> what a page will look like rendered in plain ascii, rich text, and
> full graphics at assorted window sizes and color depths?
Even that's making an assumption: these days it's worth considering how a
page will be rendered on a phone - either in a tiny display or through
speech.
--
Jon Skeet - [EMAIL PROTECTED]
http://www.pobox.com/~skeet/
------------------------------
From: Leonard Evens <[EMAIL PROTECTED]>
Subject: Re: weird: /dev/hda missing during lilo install
Date: Thu, 09 Sep 1999 02:10:48 -0500
Simon Oosthoek wrote:
>
> Hi
>
> I'm trying to install Mandrake 6 on a HP omnibook 4150. The install goes
> fine until I want to install the bootloader (lilo) on the harddisk
>
> it comes back with an error (not really specified). The mount table
> looks like this:
> /dev/root / ext2
> /proc /proc proc
> /tmp/hdc /tmp/rhimage iso9660
> /tmp/hda1 /mnt/ ext2
> /tmp/hda5 /mnt/local ext2
> /tmp/hda6 /mnt/usr ext2
>
> this is so weird! why /tmp/hda1, etc.?
> when I look there, I can't find them. They're also not in /dev/
> I've never had problems with this before (this is the 4th machine I
> install mandrake 6 on).
>
> I might be able to patch it using mkdev and some freaking, but I'd
> rather know what's really wrong!
>
> TIA
>
> Simon
I assume you found this by using an alternate screen to issue
bash commands. Then this would be a picture of the system
in the ramdisk during installation. It is not uncommon in
that situation for the devices to be called /tmp/hd*, but I
don't know why. I don't think the information in the mount
table has anything to do with why lilo wouldn't work. You
might find an error message on one of the other Alt screens.
But if you can boot from the installed system, it might be
easier to do so and fix lilo then. When you look at
lilo.conf, it may become clear what caused the problem.
Also check the partitioning with fdisk -l.
--
Leonard Evens [EMAIL PROTECTED] 847-491-5537
Dept. of Mathematics, Northwestern Univ., Evanston, IL 60208
------------------------------
From: Leonard Evens <[EMAIL PROTECTED]>
Subject: Re: Making new boot disk
Date: Thu, 09 Sep 1999 02:14:04 -0500
[EMAIL PROTECTED] wrote:
>
> I installed RedHat 6.0 with the 2.2.5 kernel and then replaced this
> kernel with the 2.2.12 kernel by creating a bzImage. Now I need a new
> boot disk that will completely load the linux OS without having to
> access the HD. I tried mkbootdisk 2.2.12, but it said it can't find
> vmlinuz-2.2.12.
> How do I make such a boot disk with the bzImage?
>
> Thanks,
>
> Sent via Deja.com http://www.deja.com/
> Share what you know. Learn what you don't.
Try
ls /boot
The number may be 2.2.12-x for appropriate x.
--
Leonard Evens [EMAIL PROTECTED] 847-491-5537
Dept. of Mathematics, Northwestern Univ., Evanston, IL 60208
------------------------------
From: Leonard Evens <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.networking
Subject: Re: NIC stops responding
Date: Thu, 09 Sep 1999 02:17:35 -0500
fred anger wrote:
>
> I'm having a problem on 2 Linux machines where the NIC seems to just
> stop responding. What's bizzare is the fix: unplug the cable from the
> hub, then plug it back in. Before unplugging the cable, the link light
> on the hub is out. Plugging the cable back in, the link light comes
> back on, and traffic flows through the NIC again. Both NICs are 3Com
> 3C905 Boomerangs. Any ideas?
>
> --
> fred anger
> http://www.triib.com/anger/
> BRING BACK DEJANEWS.COM
> 'RATE THIS' SUCKS!
>
> Sent via Deja.com http://www.deja.com/
> Share what you know. Learn what you don't.
It is most likely a problem with the cable or perhaps with the
card. Try changing the cable.
But it is possible there is a problem with the hub that the
cable connects to. Can you plug in another machine to the
same jack and see if that works okay?
--
Leonard Evens [EMAIL PROTECTED] 847-491-5537
Dept. of Mathematics, Northwestern Univ., Evanston, IL 60208
------------------------------
From: [EMAIL PROTECTED] (David M. Cook)
Crossposted-To:
comp.os.linux.admin,comp.os.linux.advocacy,comp.os.linux.hardware,comp.os.linux.help,comp.os.linux.questions
Subject: Re: Q? - best combo of linux distrib and apps for 3rd world
Date: Thu, 09 Sep 1999 07:30:14 GMT
On Wed, 08 Sep 1999 22:00:17 GMT, Allan L. Brown <[EMAIL PROTECTED]> wrote:
>- Reasonable performance on 486-33, 16 MB Ram, <400 MB HD.
>- All software free.
>- apps appropriate for NON-Technical end users need to do wordprocessing,
WordPerfect 8 might run OK on that machine with judicious choice of window
manager (no GNOME or KDE). Note that the free version of WP8 is not the
full version.
But frankly I don't think this is realistic hardware (I've never used X on
anything less than a 486/66 with 32MB RAM.). The above machine would
be a good X terminal, though.
Linux runs well on 486s, but not, unfortunately, when they are encumbered by
lots of fat GUI office software.
I suggest a Pentium class machine with 64MB RAM for "modern" (i.e. bloated)
office software.
Star Office may do OK on a Pentium with LOTS of RAM.
The other issue is HD space. I think you're going to have a hard time
squeezing what you need into 400MB.
>spreadsheet, email, web browsing, and callandering.
>- apps not dead (ie: old versions with no future or new development)
I wouldn't want to run Netscape 4.x in anything under 32MB RAM. And I haven't
tried to run it on a 486. I'd probably want to back off to Netscape 3, but
that's not being maintained.
Dave Cook
------------------------------
From: Simon Oosthoek <[EMAIL PROTECTED]>
Subject: Re: weird: /dev/hda missing during lilo install
Date: Thu, 09 Sep 1999 09:34:55 +0200
Simon Oosthoek wrote:
>
> Hi
>
> I'm trying to install Mandrake 6 on a HP omnibook 4150. The install goes
> fine until I want to install the bootloader (lilo) on the harddisk
>
> it comes back with an error (not really specified). The mount table
> looks like this:
> /dev/root / ext2
> /proc /proc proc
> /tmp/hdc /tmp/rhimage iso9660
> /tmp/hda1 /mnt/ ext2
> /tmp/hda5 /mnt/local ext2
> /tmp/hda6 /mnt/usr ext2
>
> this is so weird! why /tmp/hda1, etc.?
> when I look there, I can't find them. They're also not in /dev/
> I've never had problems with this before (this is the 4th machine I
> install mandrake 6 on).
I think I've found the reason for this...
I tried to make lilo add a dos partition without a bootable installation
to the bootloader. That must have been the error. When I looked to see
what was wrong, I got confused by the weird mount table.
after booting with a bootdisk and debugging lilo, it works now.
Simon
------------------------------
From: Leonard Evens <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.setup,be.comp.os.linux,alt.os.linux
Subject: Re: signal 11
Date: Thu, 09 Sep 1999 02:41:19 -0500
Ryan Richard wrote:
>
> I am trying to install Red Hat 6.0
>
> Every time I get to the screen where it asks me if I would like to
> create a boot disk, the install crashes. Regardless of if I choose yes
> or no, I get an "signal 11, the install exited abnormally error"
> message.
>
> I have taken this to two different machines. One is a rather new Dell
> P233 and the other a P166 clone. I have been told by Red Hat that this
> is a Hardware error. I am starting to be a little suspicious of that
> response. Is it possible that the media might be defective? Any one
> have ideas how I can proceed?
I've seen this also once before during an upgrade. But it
was transient and went away after trying again. If you've
tried it on two different machines and gotten the same result,
it would appear to be a problem with the media. Try getting
another CD. If RedHat won't replace it, order one from
Cheapbytes for a few dollars.
--
Leonard Evens [EMAIL PROTECTED] 847-491-5537
Dept. of Mathematics, Northwestern Univ., Evanston, IL 60208
------------------------------
From: [EMAIL PROTECTED] (Peter Schaffter)
Subject: rplay oddity
Date: 9 Sep 1999 07:58:08 GMT
Reply-To: [EMAIL PROTECTED]
I have an odd problem that crops up once in a while with rplay.
After ages of working just fine, all of a sudden my sound files
will start to get "holes" in them. A small portion of the file
plays, then there's a gap (about 1/2 second), then the file
continues.
Here's what's even stranger. When I encounter the problem, I've
found that by running the same file with the following two
commands, the problem fixes itself.
rplay -R 1 [filename.au]
rplay -R 0 [filename.au]
Anybody know why this is happening? And is there a more elegant
fix than the one I've just given?
Thanks.
--
PTPi
(Peter Schaffter)
------------------------------
From: [EMAIL PROTECTED] (Peter Schaffter)
Subject: rplay oddity
Date: 9 Sep 1999 08:00:22 GMT
Reply-To: [EMAIL PROTECTED]
I have an odd problem that crops up once in a while with rplay.
After ages of working just fine, all of a sudden my sound files
will start to get "holes" in them. A small portion of the file
plays, then there's a gap (about 1/2 second), then the file
continues.
Here's what's even stranger. When I encounter the problem, I've
found that by running the same file with the following two
commands, the problem fixes itself.
rplay -R 1 [filename.au]
rplay -R 0 [filename.au]
Anybody know why this is happening? And is there a more elegant
fix than the one I've just given?
Thanks.
--
PTPi
(Peter Schaffter)
------------------------------
From: brian@com
Subject: how to find font used by an xterm?
Date: 9 Sep 1999 00:21:43 -0700
hi,
I tried xwininfo and xprop, and neither want to tell me the font used
by the current xterm. I looked at .Xdefaults, and xterm*font is commented
out. I did xrdb -q, and do not see anything.
any idea?
thanks.
brian
------------------------------
** 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.misc) 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-Misc Digest
******************************
