Linux-Advocacy Digest #734, Volume #30 Fri, 8 Dec 00 06:13:04 EST
Contents:
Re: Take a conglomeration of copper and silicon and make it useful ("Tom Wilson")
Re: A Microsoft exodus! ([EMAIL PROTECTED])
Re: Linux is awful ("Adam Short")
Re: A Microsoft exodus! ([EMAIL PROTECTED])
Re: A Microsoft exodus! ([EMAIL PROTECTED])
Re: keeping a cynical gaze and a jaundiced eye on Big Brother's (SwifT -)
Re: Just in case anybody is wondering about reliability (SwifT -)
Re: I switched! No more Netscape! (Ian Davey)
Re: A Microsoft exodus! ([EMAIL PROTECTED])
Re: Linux is awful ("Adam Short")
Re: Segmentation fault (core dumped) (mlw)
Re: Linux is awful ("Anders Mörtsell")
Re: Just in case anybody is wondering about reliability (mlw)
Re: Linux is awful (Michel Bardiaux)
----------------------------------------------------------------------------
From: "Tom Wilson" <[EMAIL PROTECTED]>
Crossposted-To:
comp.os.ms-windows.advocacy,comp.os.ms-windows.nt.advocacy,comp.sys.mac.advocacy
Subject: Re: Take a conglomeration of copper and silicon and make it useful
Date: Fri, 08 Dec 2000 10:31:45 GMT
"westprog 2000" <[EMAIL PROTECTED]> wrote in message
news:90nvqv$7hd$[EMAIL PROTECTED]...
> In article <NY8X5.670$[EMAIL PROTECTED]>,
> "Tom Wilson" <[EMAIL PROTECTED]> wrote:
> >
> > "westprog 2000" <[EMAIL PROTECTED]> wrote in message
> > news:90j068$4tm$[EMAIL PROTECTED]...
> ...
> > Teaching a pupil to rely on his/her knowledge and not a compiler to
> > catch
> > mistakes isn't a bad thing at all. The first couple of tries will be
> > very
> > frustrating. You learn quick though. Speaking from experience here.
> > Besides,
> > most errors are logical and a compiler or VM can catch only the most
> > obvious. We humans are far better than they are in that area.
>
> There are many errors which will slip past even the most rigorous
> compiler - but many errors will also evade discovery through even a
> stringent testing program. Trainee programmers need to be taught this
> as almost their first lesson. They need to learn that half an hour
> spent scrutinising their code while it is still on paper/in the editor
> can save months of effort when the program has been deployed.
Very true.
> ...
> > They're made easier when a student has no option but to correct them
> > to make
> > his/her program work. Again, speaking from personal experience.
>
> When a buggy prgram passes the arbitrary set of tests prescribed, the
> programmer has the false impression that the program is correct. A look
> at the quality of most application software on the market will show the
> consequences of that attitude.
The only test, IMHO, is to give the completed copy to a bunch of folks to
Beta for me. There is no such thing as a do-all arbitrary test. The only
TRUE test is the real world. I prefer non-programmers to do my testing, BTW.
The less experience the better. Those folks can abuse software like no
others.
>
> ...
> > > But many programming faults do not bite you in the ass immediately.
> > > Sometimes you can fix them by using a logical patch.
>
> > > Example - a novice programmer runs one past the end of an array
> > >(uses a
> > > <= instead of a <). The correct fix should be in the guard for the
> > > loop
> > > - but increasing the size of the array by one will work just as
> > > well.
>
> > An example of a logical error.that won't be caught by a high level
> > compiler
> > or interpreter. The assembly student and the VB student will have to
> > resort to the exact same techniques.
>
> Which is why I favour making the programmer start by writing the thing
> out on paper, and demonstrating, to his/her own satisfaction, and the
> satisfaction of the instructor, that the thing works. That is hard
> work, if done right.
That's the way I was taught.
>
> ...
> > If you think I advocate just sitting a pupil down in front of a
> > terminal,
> > handing him/her an opcode book, and saying "OK, now, be a programmer",
> > you're mistaken. No-one said anything about ignoring the boolean
> > table,
> > top-down or modular design, loop structure, data structures, or
> > documentation practices. All I advocate is parking their fannies in
> > front of
> > a terminal to put those ideas into quick practice, ASAP.
>
> I favour giving them a pencil and paper, and a hard time.
I reserve the hard time when they're at the terminal. <g>
I still do a little training and enjoy the sarcastic raise of the eyebrow
when they have trouble at the terminal. I indulge my inner smartass at times
<g>
>
> ...
> > > No, the programmers job is to specify how data is to be extracted
> > > from
> > > a table. It is the job of the copper and silicon to take the
> > > programmers instructions and execute them.
>
> > And I'm arguing that the job gets done better and more efficiently
> > when a
> > student truly understands the machine he/she is writting for.
>
> I don't think it needs any more than a grasp of the different orders of
> speed involved in accessing random memory and sequential long term file
> storage. Someone can be a good, useful programmer with little
> understanding of the internals of the machine. They simply need to know
> the behaviour of the interface layer that they are dealing with.
I understand your argument. However, I know what getting my proverbial
fingers dirty under the hood did for me. It helps that I have some EE
training under my belt. Brings the box into focus.
>
> ...
> > Have you ever noticed the plethora of poorley written device drivers
> > today?
> > The Windows drivers in particular?
>
> I have noticed that the standard of all modern software is woefully
> low. Device drivers are no worse than the rest of the stuff.
I think a lot of it is because of reliance on programming packages, such as
VisualStudio, that allow you to code without thinking. I remember an
instructor of mine who visited and saw me using Clarion Professional
Developer. He just shook his head. I understand why, now. Once I started
hacking around with Linux using EMACS, I was shocked as to how my skills had
withered. Also, since computer time is no longer an expensive commodity like
it used to be, I feel CS instructors don't spend enough time emphasising the
"write it out first" factor. Programming is becoming more trial and error
than anything else.Though that may sound contrary to what i've said
previously, they still need to have it written out before their butts hit
the chair.
The point I was making about the device drivers, BTW, results, in my
opinion, from a lot of hardware-based ignorance. Just having an interface
description sometimes isn't enough. I've written some myself. The more you
actually understand about what it is you are talking to, the better.
I learned more about Tape Drives than I ever really wanted to...:)
>
> ...
> > > The learner programmer should be thinking about one thing - the
> > > program. Worrying about the box it will possibly run on should be
> > > left for later.
>
> > Why?
>
> Because programming is hard. It is difficult to write good programs,
> without having to learn about computers as well.
But, it isn't impossible. My programming improved by leaps and bounds the
more I delved into the hardware end of things. Understanding the medium is
important. Especially its' limitations.
>
> ...
> > > What programmers need to do when they start is to write programs,
> > > and
> > > to read them, and to see whether they will work or not. The easy
> > > availability of computing equipment has made the code/debug/fix
> > > cycle far too easy.
>
> > Good Lord, yes. I'm all for having it written out beforehand. That
> > should always be manditory.
>
> Not just written out - analysed, tested and debugged - all on paper.
> (Or using a text editor, I don't mind).
That used to be par for the course back when you were on time-share
terminals, like I was in college.
>
> > They still need to get their feet wet ASAP though. I'm
> > tired of all the hand-holding that passes for education today.
>
> I'd let a programmer access a machine at about the same stage of his
> career as a doctor is let deal with a patient.
Well, the big difference there is that, say Compaq, won't hit you with a
malpractice suit when you crash a box.
>
> > > My first programming job was on a lathe. There is quite an
> > > incentive to
> > > get it right first time under those circumstances.
>
> > LOL.
>
> > Been there. Did a lot of work upgrading and interfacing to milling
> > machines
> > and lathes. And yes, mistakes are EXPENSIVE!
>
> But noisy and entertaining. Programming a waterworks is good as well.
> Programming (or hardware) errors can result in map changes.
I knew a guy, old customer, who managed the water works in Marietta, OH. I
didn't envy him in the least!
--
Tom Wilson
Registered Linux User #194021
http://counter.li.org
------------------------------
Crossposted-To:
comp.os.ms-windows.nt.advocacy,comp.os.ms-windows.advocacy,comp.sys.mac.advocacy,comp.os.os2.advocacy,comp.unix.advocacy
From: [EMAIL PROTECTED]
Subject: Re: A Microsoft exodus!
Reply-To: [EMAIL PROTECTED]
Date: Fri, 08 Dec 2000 09:28:59 GMT
Steve Mading writes:
>>> Something being similar to itself isn't the kind of intuitive you've
>>> been talking about though,
>> But we weren't talking about "similar to itself". We were talking
>> about "thousands of OTHER electrical appliances" (emphasis added).
>>> as you pointed out when I tried bringing
>>> up vi's internal consistency. So you seem to be talking about
>>> things being similar to *other* things.
>> Perfectly consistent with "OTHER electrical appliances" (emphasis
>> added).
> Yeah, and 'vim' is highly intuative when compared to other vi
> clones. It's the same thing.
Thus doesn't count.
> If you get to define the boundry
> wherever you like, you can make anything look intuitive or
> unituitive, as needed to suit your argument, by just shifting
> the boundry around.
The key word here is "if".
------------------------------
From: "Adam Short" <[EMAIL PROTECTED]>
Crossposted-To:
comp.os.linux,comp.os.linux.x,comp.os.ms-windows,comp.os.ms-windows.advocacy
Subject: Re: Linux is awful
Date: Fri, 8 Dec 2000 10:42:26 -0000
Yes, sorry, thats what I meant.
Les Mikesell <[EMAIL PROTECTED]> wrote in message
news:QsZX5.40810$[EMAIL PROTECTED]...
>
> "Erik Funkenbusch" <[EMAIL PROTECTED]> wrote in message
> news:ZzPX5.3224$[EMAIL PROTECTED]...
> > "Adam Short" <[EMAIL PROTECTED]> wrote in message
> > news:f4KX5.9$[EMAIL PROTECTED]...
> > > From what I've read here, some people seem to be running completely
> > > different versions of Windows to anything I've ever come across. MY
> > Win98SE
> > > does exactly what Jerry says w/r to the registry. MY Win98SE install
> prog
> > > formats the Windows partition if it finds a FAT32 partition with no
> > > filesystem actually written to it, regardless of what I want it to do.
> >
> > You're making no sense here. In order for a partition to be FAT32, it
> must
> > have a filesystem since FAT32 *IS* a filesystem. If there's no
> filesystem,
> > it's unformated and has no type.
>
> I think he means a dos type partition that was created with fdisk but
> currently unused. And I think this is correct, but I don't have a system
> handy to test.
>
> Les Mikesell
> [EMAIL PROTECTED]
>
>
>
------------------------------
Crossposted-To:
comp.os.ms-windows.nt.advocacy,comp.os.ms-windows.advocacy,comp.sys.mac.advocacy,comp.os.os2.advocacy,comp.unix.advocacy
From: [EMAIL PROTECTED]
Subject: Re: A Microsoft exodus!
Reply-To: [EMAIL PROTECTED]
Date: Fri, 08 Dec 2000 09:34:53 GMT
Steve Mading writes:
>>> And incendentally, my only objection here is in your two-facedness
>>> when you say intuitiveness is all relative (to which I agree), and
>>> yet insist on making blanket statements about intuitiveness without
>>> the appropriate qualifiers.
>> You're erroneously presupposing "two-facedness", given that I haven't
>> made any blanket statements without the appropriate qualifiers.
> Hmm, I've responded to this in another part of the thread
> already.
And I've responded to your response.
> In the interest of trimming the branches, I'll not
> repeat the response here.
And I won't repeat my response either.
>>> (As evidenced by your insistence that it is possible for a part of
>>> a computer to be "intuitive" to all.)
>> Where did I allegedly make such an insistence?
> In this very thread.
Incorrect.
> Remember the power cord?
Where did I say that the power cord is intuitive to "all"? There is a
big difference between being intuitive and being intuitive to "all".
------------------------------
Crossposted-To:
comp.os.ms-windows.nt.advocacy,comp.os.ms-windows.advocacy,comp.sys.mac.advocacy,comp.os.os2.advocacy,comp.unix.advocacy
From: [EMAIL PROTECTED]
Subject: Re: A Microsoft exodus!
Reply-To: [EMAIL PROTECTED]
Date: Fri, 08 Dec 2000 09:33:16 GMT
Steve Mading writes:
>> Actually, the first thing to learn is how to exit the editor,
>> hence the ZZ.
> I'd been using vi for 1 month before I ever heard of that command,
> and I still never use it. It is not the first command you have to
> learn. Learning it as ":wq" is much better, as this is consistent
> with the other types of quitting commands.
The point is that you need to learn how to exit the editor before
other less basic commands. The actuall keystrokes used are irrelevant.
> ZZ was just a quick shortcut mapping to the 'real' command.
Why do you suppose it was added?
>>> Hit escape if you aren't sure. Now you are.
>> So much for efficiency.
> It takes roughly zero time to do that, and you only do it if
> something distracted you long enough to forget where you are.
Such distractions happen rather often.
>>> It is consistent with being the end of the type of motion command
>>> you gave.
>> Do you consider d to be a "motion command"?
> '$' is the motion command, not 'd'.
But you called it "the end of the type of motion command you gave".
> That's the way the other ones work:
> d {motion command} (delete from cursor to the moved-to-spot)
> y {motion command} (yank from cursor to the moved-to-spot)
> > {motion command} (indent block from cursor to moved-to-line)
> < {motion command} (outdent block from cursor to moved-to-line)
> v {motion command} (visually select from cursor to moved-to-spot)
> ....etc... (the last one (v) didn't exist in the original vi, but
> exists in all the new incarnations of it)
Doesn't specify whether $ is the end of line or the end of file.
------------------------------
From: SwifT - <[EMAIL PROTECTED]>
Crossposted-To: alt.2600,alt.privacy,misc.comp.os.linux.help
Subject: Re: keeping a cynical gaze and a jaundiced eye on Big Brother's
Date: Fri, 8 Dec 2000 11:39:29 +0100
On Fri, 8 Dec 2000, Jonathan Bazemore wrote:
> Linux7.0
You mean RH 7.0
> There's a service called up2date, package 2.0-4, which does the following:
>
> Description:This is a daemon which handles the task of connecting
> periodically to the Red Hat Network Services servers to check for updates,
> notifications, and PERFORM SYSTEM MONITORING TASKS according to the service
> level that this server is subscribed for.
>
> It was set to start up and run automatically, and I didn't subscribe to
> anything.
>
> Excuse me, isn't that just a long way to say back door?
I don't know RH7 enough, since I've read more than a dozen bad reviews
about it. I'll stick to RH6.2 till RH7.2 comes out. But about this
up2date-deamon:
If it is true that this deamon automatically performs system monitoring
tests without knowledge of the user, than this is very wrong. If, on the
other hand, this deamon only performs this tests when the user asks for
them (which your description says) everything is OK, since the user must
have asked explicitly about it. Also, but I'm not sure, this performance
testing only applies to people who have bought the 7.O (with build-in
support) and asked for this testing...
That the deamon checks for updates etc... is only a good thing, as long as
it doesn't automatically installs them. I hope it only sends a message to
the root that there is an upgrade for foo on the server bar.
--
SwifT
------------------------------
From: SwifT - <[EMAIL PROTECTED]>
Subject: Re: Just in case anybody is wondering about reliability
Date: Fri, 8 Dec 2000 11:33:40 +0100
On 8 Dec 2000, Jeff Silverman wrote:
> Netcraft has a service where they measure the reliability of various operating
>systems. The URL is
> http://uptime.netcraft.com/hosted?netname=DN-CIDR2,209.207.128.0,209.207.255.255
>
> Interestingly enough, of the top 50 sites in terms of uptime, EVERY ONE OF THEM is
>running Apache on
> either Linux or Solaris, or BSD/OS.
It has been posted more than enough that Win2K is too "new" on the marked.
System admins must get the feeling about it, before they can thrive it up
to good performances. Also, most sysadmins wait till servicepack 4 comes
out, since (in the line of NT) this sp will make 2K as solid as necessary.
--
SwifT
------------------------------
From: [EMAIL PROTECTED] (Ian Davey)
Subject: Re: I switched! No more Netscape!
Date: Fri, 08 Dec 2000 10:46:06 GMT
In article <[EMAIL PROTECTED]>, spicerun <[EMAIL PROTECTED]> wrote:
>Mozilla 0.6 is now on my system and working better than I had hoped.
>And definitely better than Netscape 6 or 4.76!!
Mozilla 0.8 should be out in a few weeks, 0.6 is just a release to give parity
with Netscape 6 :-)
>Mozilla ought to be nice by the time Mozilla 1.0 is released. It is
>interesting how the Headers identify this version of Moxilla as Mozilla 5.
And it will stay that way (in Netscape 6 as well). The version goes at the end
of the string.
ian.
\ /
(@_@) http://www.eclipse.co.uk/sweetdespise/ (dark literature)
/(&)\ http://www.eclipse.co.uk/sweetdespise/libertycaptions/ (art)
| |
------------------------------
Crossposted-To:
comp.os.ms-windows.nt.advocacy,comp.os.ms-windows.advocacy,comp.sys.mac.advocacy,comp.os.os2.advocacy,comp.unix.advocacy
From: [EMAIL PROTECTED]
Subject: Re: A Microsoft exodus!
Reply-To: [EMAIL PROTECTED]
Date: Fri, 08 Dec 2000 09:40:49 GMT
Steve Mading writes:
>>> Okay, I take it back. You did qualify it - but with entirely the wrong
>>> group.
>> Who or what is the "wrong group"? I qualified it for anyone who chose
>> to read what I wrote.
>>> It only looks unintuitive to users who have familiarity with
>>> other text editors first.
>> On the contrary, it can also be non-intuitive to users that have
>> no familiarity with any text editors. I did not restrict my statement
>> to just those who have used text editors, though they have the added
>> problem of "unlearning" something.
>>> This is not the same as the set of all first-time vi users.
>> Irrelevant, given that I didn't say it is.
>>> You didn't qualify it down far enough.
>> Sure I did. You simply overinterpreted what I wrote.
> Sorry, I gave you the benefit of the doubt and assumed you
> were stating something that would make sense to me.
I gave you the benefit of the doubt and assumed you would understand
what I wrote.
> The notion that someone who has never encountered any text editor
> before could still find editor A more "unintuitive" than
> editor B is an alien concept I can't fathom.
Why? Suppose you get off an airplane in a country that uses a
language that you don't understand, and you've never been there
before. Sign A has a word on it in that unfamiliar language
while Sign B has nothing but an arrow on it. Now can you fathom
the concept?
> They are both equally uninituive to someone who has had no
> experience with anything like them before.
On the contrary, the people I've worked with expect the pressing of
the letter 'k' to put the letter 'k' on the screen, not move the
cursor, or in the case of a new file, a beep.
------------------------------
From: "Adam Short" <[EMAIL PROTECTED]>
Crossposted-To:
comp.os.linux,comp.os.linux.x,comp.os.ms-windows,comp.os.ms-windows.advocacy
Subject: Re: Linux is awful
Date: Fri, 8 Dec 2000 10:55:06 -0000
> I guess he is a average Linux user, ie. Troll.....
I know you may not be used to this, but this is what's known as a
"discussion", it means to look at ideas, to put forward arguments and to
deal with them objectively. It may have started out as a slightly half-assed
observation, but there are some good points coming out here, on both sides.
A one line quip that isn't even grammatically correct, has very little place
in such a discussion.
I happen to like Linux and dislike Windows. Thats my position and I'm simply
putting forward arguments for that position. If you don't like it, at least
come up with a sensible argument.
Adam
------------------------------
From: mlw <[EMAIL PROTECTED]>
Subject: Re: Segmentation fault (core dumped)
Date: Fri, 08 Dec 2000 06:01:03 -0500
news wrote:
>
> While working on a project ( which I thought to be quite small ) I have
> received this message:
> Segmentation fault (core dumped)
> Everything compiled okay. This happened as a result of a.out
> I was wondering if someone could point out to me what I should be looking
> for that might cause such a message.
> Thanks in advance.
>
> Joarder
> [EMAIL PROTECTED]
>
> [EMAIL PROTECTED]
If you compiled the program as:
gcc program.c[pp]
do this:
gcc -g program.c[pp]
The -g option includes debug info.
Then run the program as:
gcc a.out
type 'r' and see where it takes you.
--
http://www.mohawksoft.com
------------------------------
From: "Anders Mörtsell" <[EMAIL PROTECTED]>
Crossposted-To:
comp.os.linux,comp.os.linux.x,comp.os.ms-windows,comp.os.ms-windows.advocacy
Subject: Re: Linux is awful
Date: Fri, 8 Dec 2000 12:01:04 +0100
"Adam Short" <[EMAIL PROTECTED]> wrote in message
news:143Y5.2673$[EMAIL PROTECTED]...
> > I guess he is a average Linux user, ie. Troll.....
>
> I know you may not be used to this, but this is what's known as a
> "discussion", it means to look at ideas, to put forward arguments and to
> deal with them objectively. It may have started out as a slightly
half-assed
> observation, but there are some good points coming out here, on both
sides.
>
> A one line quip that isn't even grammatically correct, has very little
place
> in such a discussion.
>
> I happen to like Linux and dislike Windows. Thats my position and I'm
simply
> putting forward arguments for that position. If you don't like it, at
least
> come up with a sensible argument.
>
If you try to run a discussion on that subject, you are very likely to have
to cope with such quips. I think that this kind of discussion is no use,
because no real Linux user will ever agree on that Windows could be useful
for some tasks and vice versa. You could go on discussing forever, and no
one will change their minds on this, it's all a matter of personal taste,
and I don't think you can change that no matter for how long you discuss it.
However I was surprised to see that it was a Windows user who put in the
"He's a troll" argument. From what I have seen, that's what linux users tend
to do when they run out of arguments, which is a petty, because I'm a Linux
fan myself.
/ Anders
------------------------------
From: mlw <[EMAIL PROTECTED]>
Subject: Re: Just in case anybody is wondering about reliability
Date: Fri, 08 Dec 2000 06:08:42 -0500
SwifT - wrote:
>
> On 8 Dec 2000, Jeff Silverman wrote:
>
> > Netcraft has a service where they measure the reliability of various operating
>systems. The URL is
> > http://uptime.netcraft.com/hosted?netname=DN-CIDR2,209.207.128.0,209.207.255.255
> >
> > Interestingly enough, of the top 50 sites in terms of uptime, EVERY ONE OF THEM is
>running Apache on
> > either Linux or Solaris, or BSD/OS.
>
> It has been posted more than enough that Win2K is too "new" on the marked.
> System admins must get the feeling about it, before they can thrive it up
> to good performances. Also, most sysadmins wait till servicepack 4 comes
> out, since (in the line of NT) this sp will make 2K as solid as necessary.
I don't know about anyone else, I heard that about the initial release
of Win2K, and we have heard the same about every release and SP of
NT/2K. This game of "The next service pack will be solid" got really old
back in NT 4.0 SP1. Does anyone seriously believe that MS has any
credibility when it comes to reliability?
--
http://www.mohawksoft.com
------------------------------
From: Michel Bardiaux <[EMAIL PROTECTED]>
Subject: Re: Linux is awful
Date: Fri, 08 Dec 2000 11:09:01 GMT
> Steve Mading wrote:
>
> > I think the reason for the hostility you see here is that people
> > like me, who *like* the CLI command set, don't want to lose it.
> > This is why there is hostility toward those who say the CLI is
> > annoying and should be superceeded by the GUI. I'm all in favor
> > of making the GUI better, but not at the expense of the CLI. In
> > an ideal world, every single thing on the computer could be
> > accessed either way, with the exception of those few apps that
> > absolutely must be GUI by their very nature and don't make any
> > sense as CLI apps (like The Gimp).
Even that is a wrong example; a superlative image-processing application
like GIMP *absolutely* needs some kind of 'batch' mode where a sequence
of operations can be automatically applied to, say, a thousand images.
--
Michel Bardiaux
------------------------------
** 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
******************************
