Linux-Advocacy Digest #669, Volume #29 Sun, 15 Oct 00 10:13:03 EDT
Contents:
Re: welcome to the world of objects (The Ghost In The Machine)
Re: Linux Sucks (Brendan Heading)
Re: welcome to the world of objects (Thomas Corriher)
Re: Claire Lynn ([EMAIL PROTECTED])
Re: Claire Lynn ([EMAIL PROTECTED])
Re: Linux Sucks ([EMAIL PROTECTED])
Re: Suggestions for Linux ("MH")
Re: Anybody want to test a widget? (2:1)
Re: Suggestions for Linux (2:1)
Re: Why is MS copying Sun??? ("Weevil")
Re: Microsoft kicked off the Web! ("Otto")
Re: How low can they go...? ("Weevil")
Re: Suggestions for Linux
----------------------------------------------------------------------------
From: [EMAIL PROTECTED] (The Ghost In The Machine)
Subject: Re: welcome to the world of objects
Date: Sun, 15 Oct 2000 12:20:43 GMT
In comp.os.linux.advocacy, Richard
<[EMAIL PROTECTED]>
wrote
on Sun, 15 Oct 2000 05:55:02 GMT
<[EMAIL PROTECTED]>:
>Thomas Corriher wrote:
>> On 10 Oct 2000 22:38:30 GMT, Steve Mading
>> <[EMAIL PROTECTED]> wrote:
>>
>> (Unless you are cynically referring to the fact that
>> >many humans eschew logic and accept illogical things
>> >(like a self-creating God.))
>>
>> Most people who believe in (accept) the existence of God,
>> would in most cases, be more logical than those who dispute
>> the existence of God. They normally don't go about attacking
>
>Isn't it wonderful how Americans can remain so thoroughly
>ignorant, clueless on the internet?
We do seem to do that often enough, don't we?
I would hope I'm one of the few who actually has half a clue
(and an interest in finding the other half), but I do wonder
about some of my compatriots.
--
[EMAIL PROTECTED] -- insert random American tourist here
------------------------------
From: Brendan Heading <[EMAIL PROTECTED]>
Subject: Re: Linux Sucks
Date: Sun, 15 Oct 2000 13:18:08 +0100
In article <[EMAIL PROTECTED]>,
[EMAIL PROTECTED] writes
>I never said any of it was a MS invention. Winamp certainly isn't. I
>am saying that Linux movement is cloning interface,applications etc
>all in an effort to compete with Windows.
I'd be interested to know, Claire : where do you think MS got the idea
for Windows NT Terminal Server Edition (now bundled in W2K Server) from
?
--
Brendan Heading, Belfast, Northern Ireland
Tús maith leath na hoibre...
------------------------------
From: [EMAIL PROTECTED] (Thomas Corriher)
Subject: Re: welcome to the world of objects
Reply-To: [EMAIL PROTECTED], abuse@[127.0.0.1]
Date: Sun, 15 Oct 2000 12:32:46 GMT
It is quite interesting how a remark which intentionally
attacked Christians on a personal level is considered
"appropriate" in a Linux group, whilst attempts to defend
Christians is considered grossly inappropriate. Of course,
the reactions of those respondents were merely "logical"
and "scientific", right? Stifling dissent with personal
attacks must be part of the "logical scientific method",
eh? I suppose I can only be logical when I accept their
religion.
I will not be silenced by those with despicable motives.
I will stand for my beliefs, and for what is noble. Some
of the audience will not like this, and no good deed ever
goes unpunished.
The malicious attacks by a few individuals demonstrate that
I am a better person than them. I have no regrets about
anything that I stated, and I intend to again speak candidly
in the future. If some people hate me for this, then I will
know with certainty that I am doing right. I do not answer
to anyone here.
The fact that no person has yet defended my logic or even
my right to dispute the attack on Christians is not a very
positive sign about the character of the Linux community.
I am honestly surprised and disappointed.
--
From the desk of Thomas Corriher
The real email address is:
corriher at bellsouth.
net
------------------------------
From: [EMAIL PROTECTED]
Subject: Re: Claire Lynn
Reply-To: [EMAIL PROTECTED]
Date: Sun, 15 Oct 2000 12:46:59 GMT
I was gone for a month and a half and this group sunk to a new low in
dull! Gore+Bush taxes, schools and several other OT that went on
forever.
Let's liven this place up a little.
A couple of people called me nasty names, so I am returning the favor.
At least you guys are exercising your fingers replying to me.
claire
On Sun, 15 Oct 2000 00:20:17 -0400, David M. Butler
<[EMAIL PROTECTED]> wrote:
>[EMAIL PROTECTED] wrote:
>
>> Whatever.
>>
>> Most of the Linvocates in this group can't advocate their collective
>> ass's out of a paper bag. They may be technically astute, but their
>> companies lock them away in rubber coding rooms far from the clients
>> because they are so, well, geeky.
>
>*giggle*
>
>You certainly like to stir up trouble in here, doncha? :P
>
------------------------------
From: [EMAIL PROTECTED]
Subject: Re: Claire Lynn
Reply-To: [EMAIL PROTECTED]
Date: Sun, 15 Oct 2000 12:48:16 GMT
On Sun, 15 Oct 2000 11:01:29 GMT, [EMAIL PROTECTED] (The
Ghost In The Machine) wrote:
>This is a generalization.
If we we in the general population it would be. We are in COLA where
things are much different
>>
------------------------------
From: [EMAIL PROTECTED]
Subject: Re: Linux Sucks
Reply-To: [EMAIL PROTECTED]
Date: Sun, 15 Oct 2000 12:52:08 GMT
They probably stole it from someone.
Where do you think Firestone got the idea to use vulcanized (I'll bet
you guys love that word) rubber in a tire, albeit with a couple of
bugs?
See number one.
claire
On Sun, 15 Oct 2000 13:18:08 +0100, Brendan Heading
<[EMAIL PROTECTED]> wrote:
>In article <[EMAIL PROTECTED]>,
>[EMAIL PROTECTED] writes
>>I never said any of it was a MS invention. Winamp certainly isn't. I
>>am saying that Linux movement is cloning interface,applications etc
>>all in an effort to compete with Windows.
>
>I'd be interested to know, Claire : where do you think MS got the idea
>for Windows NT Terminal Server Edition (now bundled in W2K Server) from
>?
------------------------------
From: "MH" <[EMAIL PROTECTED]>
Subject: Re: Suggestions for Linux
Date: Sun, 15 Oct 2000 09:07:12 -0400
> > The no-life cultists who enjoy being intitiates into the secret
knowledge
> > of UNIX internals could still play with the "hard" version, and make
snide
> > comments about those too weak to use it.
These users will soon be relegated to cola and college computer labs.
But, the snide comments will persist, as this is how they use Unix as some
sort of self validating computer-superiority litmus test. If you're sitting
there doing your homework on Word, or debugging in the VB ide, you're
branded an idiot by these glaring examples of arrested development.
Sad as hell IYAM.
------------------------------
From: 2:1 <[EMAIL PROTECTED]>
Subject: Re: Anybody want to test a widget?
Date: Sun, 15 Oct 2000 15:07:10 +0100
[EMAIL PROTECTED] wrote:
>
> In article <[EMAIL PROTECTED]>,
> 2:1 <[EMAIL PROTECTED]> wrote:
> > [EMAIL PROTECTED] wrote:
> > >
> > > In article <[EMAIL PROTECTED]>,
> > > 2:1 <[EMAIL PROTECTED]> wrote:
> > > >
> > > > Sorry i can't help, though.
> > >
> > > Rat bastard.
> >
> > When I start on GTK and have some time I might help. I understand your
> > frustration, though.
>
> S'alright. Just kidding around.
>
> > I understand your frustration, though.
>
> Uhhh.... S'alright. Just kidding around. You're skipping, by the way.
Never leave the keyboard half way through a message... :-)
> > Anyone want to help on a FORTH project? It's a nice, fast, portable (I
> > hope)implementation designed to be embedded very easily in C programs.
>
> What's it do? Just curious.
Forgive me if I over-explain a little, but I don't know how much you
know.
Its a forth implementation, so it has 2 stacks: a main stack (which
deals with all the calculations) and a return stack (for loops, function
calls, etc, etc).
It is a very weakly typed language. The main stack is a union of what
ever data you may want to store (by defaault, a union of signed long,
float and void *). Esentially, you have to keep track of what is on the
stack, although this is not actually too hard.
Because it is a stack based languagem it is RPN and has no parentheses.
So
2*(3+4) becomes 3 4 + 2 *
All operations on the stack are done by the push, pop, peek and replace.
The last too are for speed, really.
Commands can take the form of variables, constants, c functions or
definitions. A definition is a collection of commands, like a function.
New commands (in C) are of the type:
(void)fcn*(parser)
where parser is a pointer to a data type containing all the relavent
information for the parser.
A simple command might be (to add the 2 numbers on the main stack and
replace them with the sum.
void add(parser info)
{
push(info, pop(info).i + pop(info).i);
}
Before it is first used, it must be registered with the parser:
c_reg(info, "+", add);
from now on, when ever the parser encounters a +, it will add 2 numbers.
All commands are added in thios way. There are no built in commands,
even the ones that come with it as default. The only built in response
is that if it reads something that isn' recognised as a command, and is
then recognised as a number, it puts it on the stack.
The minimal program using this (analogus to TCLsh, if my understanding
is correct) is:
main()
{
parser info;
init_parser(info)
parse_file(stdin, info);
}
This would read commands from the standard input and run what ever
programs were there.
As you can see, I have designed it the be integrated in to a C program
with minimal possible effort.
Its history is as follows:
Origionally, the RPN parser was written for a dialog box generator for
(currntly) text mode dialog boxes. The RPN parser was needes because I
wanted to have complex dependences between items in the dialog box (eg,
greying out and hiding).
The RPN parser was altered slightly, to give me a very simnple
parametric graphics file format (for my first XLib app (unfinished, of
course)), for a game. It then got waylayes, and started becoming a fully
fledged imbeddable language.
The work left now, is to make more `built in' functions (more loops,
conditionals, etc). I also want to create a toolkit or 2 ( the Xwindows
one for simple parametric wil be the first), so that it can be really
useful.
That is a very brief introduction.
If you want more info, email me.
There isn't a website for it yet, but there may well be one soon. If you
want, I'll email you when its avaliable.
If you want the sourcew code, also email me.
It'll probably go under the LGPL.
-Ed
--
Konrad Zuse should recognised. He built the first | Edward Rosten
binary digital computer (Z1, with floating point) the | Engineer
first general purpose computer (the Z3) and the first | u98ejr@
commercial one (Z4). | eng.ox.ac.uk
------------------------------
From: 2:1 <[EMAIL PROTECTED]>
Subject: Re: Suggestions for Linux
Date: Sun, 15 Oct 2000 15:26:44 +0100
> >3) We need to add superior functionality to the Linux GUI, like
> > the "Halflife" game, with openGL and 3-D icons for linux functions-
> >
> > a) A restaurant. F'rinstance, you boot linux, and you see a first
> > person view of yourself walking into a restaurant. You sit at a
> >table,
> > and tux the penguin walks over and hands you a menu. The menu has
> > linux programs grouped on pages with clickaable tabs. You click a
> >tab for
> > say, graphics, and a page turns to all the graphics programs . You
> >click
> > on a menu selection to start up the corresponding function.
> > b) An office building. You find yourself walking down a hallway,
> > each door leads to either a room or another hallway. Rooms are
> > directories with representational 3-D icons for files (like a TV
> >for viewing
> > animations, or a filing cabinet full of documents, each of which is
> >
> > a spearate manilla folder). Hallways are directories of
> >directories.
> > We could produce a tool like a .wad file editor to allow users to
> > customize the 3-D environment.
>
> You've got to be kidding me.
I remember discussing something like a doomlike file manager with my
brother. Shoot to delete, etc, but it never got this far |
|
\|/
>
> I've fantasized (and did little else) regarding a DOOM-type environment
> which would allow one to walk into various rooms (directories) and
> pick up stuff (files) and drop them in other directories, or into
> executable rooms. Read-only directories would have all the crates
> locked down; directories without search capability would cage the
> user so that he can see, but not explore; directories without read
> capability would have the lights turned off, but the user would
> have a Thingy that would allow him to type in the name of an
> object; symbolic links are transporters or strange-looking
> crates with whirly effects (broken links kill you or are a nasty-looking
> dark pool in a crate);
But if it's not a borken link, the process will only get telefragged :-)
> executables are doors to little rooms with a
> display (an X server) on one wall and a few controls on another
> ("RESTART PROCESS", perhaps); to manipulate the windows on that
> display, use the mouse on the display wall; to type, point the mouse
> at the keyboard, then start typing on your keyboard.
Corruped areas of the hard disk can have flickering lights and floors
that do damage (unless you have a rad suit, of course).
>
> Or one can shoot the keyboard, display, or control area with
> a handgun, a shotgun, or a rocket launcher, which kills the process
> with various signals and kick one back into the central process
> dispatch area, which is a special room, probably /proc, or maybe
> /proc/processes if anyone ever gets around to implementing that.
>
> Keyboard gun shoot = SIGINT;
> control gun shoot = SIGQUIT;
> display gun shoot = XKillClient();
> shotgun blast = SIGTERM;
> rocket blast = SIGKILL.
ROFL! What does the BFG9K do then? patch /proc/kcore? to remove all
trace of the process?
> Hope your benefits are paid up! :-)
>
> The mountpoints might be in open sky, in a nice green, grassy field,
> maybe with a bunny rabbit's head staked in the ground. :-)
> Or one can start in the dungeons at one's home directory, which would
> be a different color perhaps to distinguish it. A nice homey yellow,
> maybe, with a clock on the wall and a little doggie, and a desk
> in one corner with an "In" and "Out" basket. (Anyone remember BOB?)
>
> Others have actually implemented (rather silly, IMO) DOOM-type "games"
> where each process is mapped into a sergeant (complete with
> floating PID); to kill a process, one shoots the sergeant. One
> drawback is that the DOOM code occasionally has sergeants shooting
> at each other. It's a bug.
I have to have this. Where can I get it?
It wasn't a bug. Sargents were meant to shoot at each other, in DOOM.
You could have the lot really. Caco-daemons and of course, the
cyber/spider-daemon as the mother of all daemons (would that be init?).
Zombies could be zombie processes etc etc.
> Still others -- Xerox? -- have had hypothetical concepts such as
> hanging directory trees, movable walls, and such for a while;
> these might be nice visual aids, but that's about it.
>
> To all this, I say "What's the point?" What's really happening are
> magnetic domains -- or, if you prefer, bits -- are being moved around
> in an organized way. Maybe I'm old school, but I prefer
>
> "mv afile ../bdir"
>
> to
>
> "go into that room there, walk to the crate with the 'afile' sticky on it,
> and click on it, opening it. Now grab it.
>
> You now have it in your inventory. Now go back to the lift
> and push the 'Up' button. You should see another room; locate
> the door named 'bdir', walk though it. Now drop the 'afile' crate there."
Only pausing to blash the hell out of the other users on the system, if
they were trying to access the same file.
> More intuitive, perhaps, but it gets tedious after awhile, I would think!
> Although it might appeal to the gaming set, especially if there's
> a wandering Glitch monster which zaps things at random.
> (Yeah, that'll really help the seccy get *her* work done!)
>
> There's also the issue of grabbing a door. Hmm...what's wrong
> with this metaphor, eh? :-)
>
It sounds fun, for about 5 minutes. Just browsing files could give you a
game of doom, if you are on a multiuser system. Now that sounds like
fun.
IMO Quake would be better (more configurable), but less atmosphere.
-Ed
--
Konrad Zuse should recognised. He built the first | Edward Rosten
binary digital computer (Z1, with floating point) the | Engineer
first general purpose computer (the Z3) and the first | u98ejr@
commercial one (Z4). | eng.ox.ac.uk
------------------------------
From: "Weevil" <[EMAIL PROTECTED]>
Crossposted-To:
comp.os.ms-windows.nt.advocacy,comp.arch,alt.conspiracy.area51,comp.os.netware.misc,comp.protocols.tcp-ip,comp.lang.java.advocacy
Subject: Re: Why is MS copying Sun???
Date: Sun, 15 Oct 2000 08:39:54 -0500
Mike Byrns <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
>
>
> Weevil wrote:
>
> > In fact, having the source for every function called by the edit control
> > won't get you there, if what you're aiming for is binary compatibility.
> > That's what WINE is after, as you must know.
>
> Youe just need a PE loader shell to load the code from the .exe file. It
starts
> WinMain() and passes the proper parms. WinMain registers the window class
> (which you can stub) and creates the main window (which you borrow from
linux).
> You then produce a "handle" to that window mapped from however linux IDs a
> window. You then show the window using the linux function calls to do the
> same. Then you enter your message loop calling GetMessage until it
returns
> false and passing that msg struct to Translate and Dispatch Message funcs.
When
> you get false from GetMessage you terminate. The only tricky part is that
you
> have to have a consistant message based OS to pass messages to your
WindowProc
> so it can do useful things like create a client area sized standard edit
control
> when it gets the WM_CREATE message or to PostQuitMessage when it gets a
> WM_DESTROY. It's all very simple and logical.
My god! It's so simple! Over 300 Wine developers in more than a dozen
countries...how silly are THOSE guys going to feel when they see this?
I guess that's it, then. As soon as somebody alerts the Wine guys as to
your above paragraph, they can call it Wine 1.0, stamp DONE on it, and move
on to Win2k or something.
Boy, and all this time they thought there was so much more to it than that.
> > Recreating the functionality of a well-defined interface is
time-consuming,
> > but not extroardinarily difficult. Even recreating the functionality of
> > Win32 would not be hideously difficult, if its functionality were
actually
> > well documented.
>
> You really don't have any idea of just how well documented it is. Please
goto
> (sorry :-) http://msdn.microsoft.com/library/default.asp and start looking
up
> the functions I've mentioned. They are absolutely meticulously
documented.
Dang, is anybody writing this stuff down for the Wine guys?
> Many times better than their Qt, GTK or *gasp* raw X counterparts.
Well...not exactly. You can try to explain an elephant to someone all day
long and even do a really good job of it so that he has a really good idea
of what it is, but there is no substitute for showing him the elephant
itself.
Linux developers show the elephant. (That would be the source code, Mike.)
> > A ton of work, yes, but not beyond the resources of any of
> > dozens of software companies. Of course, merely recreating its
> > functionality would not allow windows apps to run on your version of
> > windows -- they would all have to be recompiled on your system.
>
> Nope. Sorry. Bzzz. You just need to read PE. It's just Intel machine
code.
> WINE doesn't run on non-x86.
WINE doesn't run on non-x86 because it does not attempt to duplicate every
single Windows DLL in existence, which is what would be required to port it
to a foreign architecture. Wine actually uses a lot of the DLLs in their
native form.
And yes, sorry, but with source-level compatibility, you do need to rebuild
the software on the emulator. You seem to be confused about what is meant
by "source compatibility." Go look it up somewhere. Perhaps MSDN can help
you.
Wine has that too, by the way. I hadn't realized they were going after
binary *and* source compatibility. Cool beans.
> > That's not what the WINE team is after. They want your existing windows
> > programs to run as is. Immeasurably more difficult than source level
> > compatibility (which is all you had your students doing). Even more so
> > since the windows API is neither thoroughly nor accurately documented.
>
> It's better documented than the Linux and Mac APIs. Sorry.
>
I don't know about the Mac, but without cracking a book on Windows or Linux,
I can guarantee that all Linux APIs are infinitely better documented than
their Windows counterparts. The ultimate documentation, of course, is the
source code.
I guess this assumes that you can read and understand the source, though.
Even without the source, I've found the man pages to be better, more
complete, and more accurate than Microsoft's documentation. But part of
this could be just a matter or taste.
> > By the way, the payoff for any company even achieving source level
> > compatibility with Windows is potentially gigantic. Plenty of companies
> > achieve something very close to perfect source compatibility with each
other
> > in Unix. Lots of Unix vendors out there. Why only one Windows vendor?
>
> Because it's copyrighted? :-)
No. :-) There are no legal barriers to developing a Windows work-alike
system, whether you're after source or binary compatibility. Anyone can
legally do this. (Didn't you wonder how the Wine team was getting away with
it?) Windows owns copyright on their source code, not on what their source
code accomplishes.
Do a little research, and when you've discovered that you're wrong and that
there is no legal barrier to doing such a thing, come back and consider this
question again.
Why is there only one Windows vendor? It is by far the largest software
market in the world.
jwb
------------------------------
From: "Otto" <[EMAIL PROTECTED]>
Crossposted-To:
comp.os.ms-windows.advocacy,comp.os.ms-windows.nt.advocacy,comp.sys.mac.advocacy,comp.unix.advocacy
Subject: Re: Microsoft kicked off the Web!
Date: Sun, 15 Oct 2000 13:43:10 GMT
"Charlie Ebert" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
: And this might be where RED HAT comes in.
: RED HAT helps to keep you from having RED ASS.
:
: http://24.94.254.33/Linux/intro.html
Interesting point of view at the above link, not to mention your
interpretation of the article. However, there are some perspective which
isn't mentioned on the site and you might've drawn the wrong conclusion.
The masses rather have "RED ASS" than "RED HAT" for the time being. This
might change by year 2005, but that would require couple of things to
happen, they are actually related to one and other and somewhat overlooked
in the article.
Linux need to become as easy to use as Windows, or the masses will be
proficient to use Linux. The former one already started, to the dismay of
hardcore Linux advocates, Linux will be controlled by the GUI instead of the
CLI. Only time will tell how successful this will be, anyone's guess is
valid on this subject at this time, including the author of the article.
As for the masses becoming proficient in the present form of Linux, good
luck. The masses have no interest in the inner working of the OS. Microsoft
built, or exploited their business on this fact.
Otto
------------------------------
From: "Weevil" <[EMAIL PROTECTED]>
Crossposted-To:
comp.lang.java.advocacy,comp.os.ms-windows.advocacy,comp.os.ms-windows.nt.advocacy
Subject: Re: How low can they go...?
Date: Sun, 15 Oct 2000 09:07:56 -0500
<[EMAIL PROTECTED]> wrote in message news:1058s8.1kb.ln@gd2zzx...
> In article <NIsE5.50444$[EMAIL PROTECTED]>,
> "Weevil" <[EMAIL PROTECTED]> wrote:
> >
> > Bill Gates belongs in jail. So does Steve Ballmer. So do a dozen other
MS
> > executives past and present. Microsoft was built on theft, lies, and
> > plagiarism, and it hasn't changed a damn bit. It never will. You're
one of
> > the apparently endless number of fools who can't or refuse to see this.
> >
> > Software in general, but especially OS software, is at LEAST 10 years
behind
> > where it would have been if not for Bill Gates, Steve Ballmer, et al,
and
> > their utterly evil willingness to stifle any and all competition
illegally
> > while investing next to nothing in improving their own products.
> >
> > This is not conjecture. It is not bitter spewing from a Microsoft
hater.
> > It is fact. It has been proven. It is reality.
>
> This is what I cannot understand. It is clear from the trial that
> microsoft personnel committed purjury but I hear nothing about
> them being charged with anything. Is there any justice in the
> USA?
See claire's reply above.
There's lots of justice in the USA, but who can afford it?
jwb
------------------------------
From: [EMAIL PROTECTED] ()
Subject: Re: Suggestions for Linux
Date: 15 Oct 2000 10:09:36 -0400
Chris Sherlock <[EMAIL PROTECTED]> spewed this unto the Network:
>unicat wrote:
>>
>> At the risk of providing fuel for the Wintrolls, I am posting some
>> friendly criticism of Linux and the Gnome/KDE GUIs.
>> (Notice to Microsoft, you can't copyright any of these ideas,
>> I am hereby copyrighting c2000
>> them and placing them in the public domain)
The ideas would have to be patented, and that costs money.
>> Linux will never complete its dominance of the computing world
>> if we are simply striving to be "as good as" Windows. We must set
>> our sights on an OS that is not just more reliable, but much much
>> easier to use than MS if we are going to see continued adoption of Linux
>Well, Linux is not just trying to be '"as good as" Windows'. IMHO in
>most areas Linux already *is* as good as Windows, anyway!
Linux isn't "trying to be" anything. It's a piece of software, not a person.
>> 2) We need to completely eliminate the command line interface.
>> That's right. Get rid of it. Anything that can't be done from a
>> GUI isn't worth doing. Remove ed,vi,emacs,vim, telnet, rlogin, rsh,
>> and especially getty from the distribution package completely.
>> Run ppp on all serial lines by default. PCs are cheaper than VT100s,
>> and we can use X-windows over ppp instead of curses. To
>> replace telnet and rlogin, use an http link and HTML pages that
>> use cgi to run commands.
>You crack me up! IMO the CLI is extremely useful. When I had to check
>servers, the CLI was what I used. Getting rid of the CLI is only
>*reducing choice*! Many people like ed, vi, emacs and vim. Why get rid
>of them?
>Running X-Windows over ppp instead of curses? I do beleive that their
>are apps that do this already.
X-Windows itself can do it. Once a PPP interface is up, the OS sees it
as another network device. _ALL_ X apps will display their graphics
on the X display specified in the DISPLAY environment variable, with
NO exceptions. If DISPLAY specifies a server that is not on the local
host, then the X app will connect to the appropriate server over the
network, and the kernel will prevent it from having to know anything
about the PPP device (which might really be an Ethernet device-- it
makes no difference to the X app).
However, X is noticably slower even over a 10Mbit connection than it is
over a Unix domain socket. Over a 38400 bps serial connection, you wouldn't
even be able to use Netscape. Running X over PPP instead of curses is
not a good idea.
>You're suggestion to use web pages for telnet sessions, etc, has already
>been implemented. I saw this in one of my sys admin magazines that I
>regularly buy. Because of the way that http is designed, I beleive that
>this idea, although it has it's merits, is rather daft. http is not
>interactive, and it does not save state! Why bother?
Using CGI to run shell commands instead of using an xterm to type
the command directly into the shell is a ridiculous idea. Not only
is HTTP non-interactive and it doesn't save state, but running
commands this way is a lot slower than running them directly in
the shell.
--
Microsoft Windows. The joke that kills.
------------------------------
** 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
******************************
