Linux-Advocacy Digest #349, Volume #26            Wed, 3 May 00 03:13:04 EDT

Contents:
  Re: The Dream World of Linux Zealots (The Ghost In The Machine)
  Re: Dvorak calls Microsoft on 'innovation' (John Jensen)
  Re: Is the PC era over? (JEDIDIAH)
  Re: Is the PC era over? (Andrew Carpenter)
  Re: My question has still not been answered.Dance..Dance...Dance... (JEDIDIAH)
  Re: Is the PC era over? (The Ghost In The Machine)
  Re: Dvorak calls Microsoft on 'innovation' ("Luke Webber")
  Re: Dvorak calls Microsoft on 'innovation' ("Luke Webber")
  Re: Are we equal? (abraxas)
  Clearing things ("Alberto Trillo")
  Re: Is the PC era over? (abraxas)

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

From: [EMAIL PROTECTED] (The Ghost In The Machine)
Subject: Re: The Dream World of Linux Zealots
Date: Wed, 03 May 2000 05:40:29 GMT

In comp.os.linux.advocacy, [EMAIL PROTECTED] <[EMAIL PROTECTED]>
wrote on Wed, 03 May 2000 02:56:30 GMT <8eo4gn$lh2$[EMAIL PROTECTED]>:
>In article <[EMAIL PROTECTED]>,
>  [EMAIL PROTECTED] wrote:
>[snippo grande]
>> Windows is a much, much better choice.
>
>No, it isn't.  I tried it, it didn't work.  And it cost too much, and
>there was no support.  Besides, the manufacturer has been found guilty
>of violating the law, and I don't support lawbreakers.

It may depend on what one's definition of "better" is.

Windows, for example, is definitely better at crashing :-),
but it also has more functionality at the base OS level.
(Whether this buys anything, I'm not sure.  WaitForMultipleObjectsEx()
in particular is a peculiar routine; every other OS I've worked
on uses fixed-length bit sets for synchronization, or builds
structures on top of said fixed-length bit sets.  Unix in particular
uses select(); the Amiga had 32-bit longwords, and I think
VAX/VMS did, too.)

Windows also has more apparent interoperability.  (It becomes
very quickly apparent that this apparent interoperability is only
applicable to -- other platforms running Windows.  OE stands for
Oh Ecch! when it comes to newsgroup posting, for example. :-) )

And Windows has a uniform low-level GUI interface, where a menu
is a menu, a button is a button, and a scrollbar is a scrollbar.

At least it used to.  The innovations I'm seeing lately are getting
downright *peculiar*; how often does one need to move the menu bar
off the top of the window, for example, or double-click on it to make
a separate area?  And then there's the QuickTime idea of mimicking
a physical device; never mind that there's no thumb to manipulate
the "thumbwheel", so one has to press and hold a button on the mouse
instead while moving it around.  It might look pretty, but the
interaction is not all that clear.  Of course moving a mouse
around is like trying to draw with a soap bar, anyway; even a trackball
(which I prefer at home) isn't perfect.  The single best implement
for drawing is the infinite-RAM number 2 Pencil with eraser,
or a reasonable facsimile thereof.... :-) [*]

At least the semitransparent icons used when moving icons (which
represent files or directories) around are somewhat useful.  I'll
give them partial credit for that; of course, they lose it when
the user is required to either figure out whether the file will
be copied, moved, or shortcut-linked to depending on various factors
(is it an executable?  Are the two windows on the same volume?
Does the source directory have write permissions?  Is the moon
waning gibbous?  Does the user have facial hair and long underwear
on?  Has Fermat's Last Theorem been solved yet??), or whether the
user is pressing a shift button -- with the other hand, no less.
(Suppose he hasn't got one?)

And cutting/copying and pasting of files is useful, but weird.  Usually,
when cutting text, the text vanishes.  However, when cutting files,
the files don't vanish until the move occurs.  Yep, that's consistent --
not.  (To be fair, it does make the operation a little more
cancelable and users probably a little less queasy.)

If you want more details on some of these, try
http://www.iarchitect.com/mshame.html
.  Many, um, "innovations" provided by Windows are noted there.

Yeah, Windows really is better...at confusing the user.  I'm not
sure about X GUIs, though, but at least with a CLI I can type in
a command and have a reasonable chance at predicting what's going
to happen next.

(Oh, Apollo DOMAIN!  Why did you have to die off and leave us with
this pathetic crud??)

>
>Sent via Deja.com http://www.deja.com/
>Before you buy.

[*] Some artists might take issue with me on this, and I for one
    can't draw a face, figure, or diagram to save me anyway. :-)
    But there are touch-sensitive art pads for the PC out there,
    AFAIK; I could easily see professional artists manipulating
    these as though they were airbrushes, charcoal, or even
    watercolor, with the right software.

-- 
[EMAIL PROTECTED] -- insert random misquote here

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

From: John Jensen <[EMAIL PROTECTED]>
Crossposted-To: comp.os.ms-windows.nt.advocacy,comp.lang.java.advocacy
Subject: Re: Dvorak calls Microsoft on 'innovation'
Date: 3 May 2000 05:42:22 GMT

Jen <[EMAIL PROTECTED]> writes:
: On Tue, 02 May 2000 19:01:51 GMT, [EMAIL PROTECTED] wrote:

: >Dvorak also charges that Microsoft hires PR agencies to make
: >pro-Microsoft postings on on-line forums (which presumably includes
: >c.l.j.a)

: Funny how he didn't name names.  How convenient.

It would be soooo entertaining to cross-check such a list with deja.com
histories.

I'm afraid though, given random variation between individuals, it is
statistically likely that *any* viewpoint will have amateur advocates.

I wonder (if we could examine the record) if these PR agencies would be
capable of generating any signal, or just contributing to noise?

John

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

From: [EMAIL PROTECTED] (JEDIDIAH)
Crossposted-To: comp.os.ms-windows.nt.advocacy,comp.lang.java.advocacy
Subject: Re: Is the PC era over?
Date: Wed, 03 May 2000 05:42:19 GMT

On 3 May 2000 05:28:19 GMT, abraxas <[EMAIL PROTECTED]> wrote:
>In comp.os.linux.advocacy Andrew Carpenter <[EMAIL PROTECTED]> wrote:
>
>> I was most disappointed when I found out about the SunRay. No matter how
>> they word it, it seems to be nothing more than a dumb terminal, plain
>> and simple. 
>
>Thats kind of the idea.  Its essentially an X-terminal.
>
>> It does no local processing beyond drawing pixels, making it
>> the equivalent of a Citrix client in hardware. it's biggest feature is a
>> smart card reader so it can do 'hot desktops'; Metaframe could almost
>> have that now. If it was a dirt-cheap appliance, maybe, but it's not:
>> $500 US plus a dedicated server.
>
>Thats actaully very cheap for X-terminals, which in the past (last time
>I checked was last year) have gone for between 1000 and 3000 dollars-US.

        This isn't 1989.

        Complete systems with their own disks can be had for that much.
        Even $500 for a mere Xterminal is a bit much. Even that much for
        a machine capable of it's own computation and some disk caching
        would be pushing things a bit.

        This is likely why NC's never caught on. The pricetag is more
        impressive to most (even PHB buyers) than some TCO figure.

>
>> This is not where NCs should be going; the idea of a 'Network Computer'
>> was to have local processing power, but remote data storage. 
>
>X-terminals are not the idea of a 'Network Computer'.  They're the idea
>of a remote-viewing station.  

        ...it depends on who you ask.

        Although, 'remote viewing station' isn't necessarily what
        an X-terminal would be as implemented as the past either.
        Unix always had the flexibility of being a fatter or thinner
        client depending on individual tastes.

[deletia]

-- 

                                                                        |||
                                                                       / | \
        
                                      Need sane PPP docs? Try penguin.lvcm.com.

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

From: Andrew Carpenter <[EMAIL PROTECTED]>
Crossposted-To: comp.os.ms-windows.nt.advocacy,comp.lang.java.advocacy
Subject: Re: Is the PC era over?
Date: Wed, 03 May 2000 15:20:33 +0930

abraxas wrote:
> 
> In comp.os.linux.advocacy Andrew Carpenter <[EMAIL PROTECTED]> wrote:
> 
> > I was most disappointed when I found out about the SunRay. No matter how
> > they word it, it seems to be nothing more than a dumb terminal, plain
> > and simple.
> 
> Thats kind of the idea.  Its essentially an X-terminal.

Even an X-terminal is one step up from this concept, at least in
functionality stakes. X-terminals do all the windowing (and
widget-handling?) locally. The SunRay is purely a screenscraper, is it
not?
 
> Thats actaully very cheap for X-terminals, which in the past (last time
> I checked was last year) have gone for between 1000 and 3000 dollars-US.

I've never worked out why... I don't really know enough about their full
capabilities. I do know some are capable of actually becoming NCs in
their own right (applications can be natively compiled and run locally).
 
> X-terminals are not the idea of a 'Network Computer'.  They're the idea
> of a remote-viewing station.

That's right... so why should the 'wave of the future' be the equivalent
of an older technology?
(X-terminals are actually *closer* to a real NC than the SunRay!)
 
> > I'd argue NCs are not just the way we'll ultimately be going, they are
> > inevitable. There will always be servers of course, and there may still
> > be PCs; but they will probably start to take on more of a role as a
> > local server to all your other networked devices, rather than a
> > workstation in itself.
> 
> Which is pretty much what an X-terminal does.  The technology for this
> has been around since the Xwindow system.  Its simple, straightforward,
> efficient (even moreso now than its ever been) and with recent talk of
> using an embedded system (linux?) to run an ssh client locally to pipe
> all your stuff, quite secure.

An X-terminal can't serve applications or files. Or do you mean the
X-terminal is the device?

If anyone doubts NCs are still on the cards, note the attention device
networking is getting from all directions; see Jini, UPNP etc. If the
peripherals can all serve on the network independently -- printers,
scanners, even hard drives -- then it is *inevitable* the end result
will be the NC as the general purpose computer.

Andrew
[ opinions are my own ]

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

From: [EMAIL PROTECTED] (JEDIDIAH)
Subject: Re: My question has still not been answered.Dance..Dance...Dance...
Date: Wed, 03 May 2000 05:52:31 GMT

On 3 May 2000 00:21:25 -0500, Leslie Mikesell <[EMAIL PROTECTED]> wrote:
>In article <[EMAIL PROTECTED]>,
>JEDIDIAH <[EMAIL PROTECTED]> wrote:
[deletia]
>
>It really is a yes or no question.  Turning it on activates NAT,
>a DHCP server on the NATed range (which it figures out), and
>demand dialing, automatically triggered by client access.  And

        What if you don't use a serial modem? What if you'd
        rather use the serial modem for something else?
        What if you would rather bridge and use the upstream
        dhcp server instead? What happens if you want to use
        2 NICs and isolate routable packets from non-routable
        ones? Does Win98SE get confused if you have no modem?

>maybe some other things I haven't found yet.  I was amazed
>that on my dual-boot machine it worked just the same as my
>hand-built setup under Linux down to the way I had set up
>dhcp.  Don't compare this to anything prior to SE - they
>got the defaults right this time, although they should warn
>you about running the dhcp server since it would destroy a 
>network that was already using one.

        That's a rather big gotcha.

        ...puts a new spin on Cable Modem provider DHCP server problems.

-- 

                                                                        |||
                                                                       / | \
        
                                      Need sane PPP docs? Try penguin.lvcm.com.

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

From: [EMAIL PROTECTED] (The Ghost In The Machine)
Crossposted-To: comp.os.ms-windows.nt.advocacy,comp.lang.java.advocacy
Subject: Re: Is the PC era over?
Date: Wed, 03 May 2000 05:53:37 GMT

In comp.lang.java.advocacy, [EMAIL PROTECTED] <[EMAIL PROTECTED]>
 wrote on Wed, 03 May 2000 03:03:49 GMT <8eo4ud$ll3$[EMAIL PROTECTED]>:
>Subject: Is the PC era over?
>
>$DEITY, I hope so.

If one means "political correctness" with the term PC, I agree with you.

If one means "personal computer", in the general sense, I
certainly hope not; I happen to like the idea of using a very
general, powerful tool to post news, for example. :-)
Not to mention run the latest 3-D gaming software, do my taxes,
and chat on the Internet.

If one means "started out as ISA-based, 8088 microprocessor, 640k is
good enough for anybody, but ultimately ended up as a little brother
with gigabytes of flat address space, but poor throughput",
then I for one am somewhat ambivalent, although I do agree that the
"PC" of today is a far cry from the PC-XT of yesteryear
(the 386 supported a 32-bit flat address space, and modern Pentium-based
"PC"'s can now carry enough RAM to use a good chunk of it).
But the Amiga was definitely more interesting (heck, so was the
Atari), although more expensive; the Apollo DOMAIN had a better
GUI (but then, it was an expensive workstation); other platforms
arguably have higher I/O capability (I can't say from personal
experience).

I do note that, for some reason, I hold less animosity towards
Intel than I do towards Microsoft.  One issue may simply be that
Intel's microprocessor isn't the only component on the "standard
motherboard" in today's PCs (presumably, the I/O throughput problem
is because of the entire motherboard architecture, not just the micro).
I do consider the 680x0 family a superior design, although the
Intel Pentium now probably outperforms Motorola's top end 680x0.
However, Motorola still has the PPC... :-)

Sigh.  Yet Another Demonstration As To Why Superior Marketing
Wins Out Over Technological Innovation If One Isn't Real Careful. :-)

>
>Sent via Deja.com http://www.deja.com/
>Before you buy.

-- 
[EMAIL PROTECTED] -- Windows.  Where do you want to go today?
                    No, you don't want to go *there*....

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

From: "Luke Webber" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.ms-windows.nt.advocacy,comp.lang.java.advocacy
Subject: Re: Dvorak calls Microsoft on 'innovation'
Date: Wed, 3 May 2000 16:23:18 +1000

The Ghost In The Machine wrote in message ...
[snip]
>There is none, there.  However, innovations abound elsewhere.
>X (The X Window System, that is, but I'll just call it X
>for the sake of brevity :-) ), for instance, was "innovated"
>in the mid-80's; TCP/IP was created in the early to mid-80's
>by the Woolongong group, IIRC.


Cripes I hope not! No, TCP/IP originated at Berkely. The Wollongong Group
just put out a commercial implementation called something like
Pathway/Access. It always struck me as being very expensive for what it was,
especially since it sprang out of work done at the University of Wollongong
(that's an industrial city south of Sydney, BTW).


[snip]
>Hardware for Unix has always been a problem.  One issue with Unix
>is that it bangs on hardware very hard (Windows 9x merely taps it
>more lightly), because of the inherent multitasking/multiuser nature
>of Unix, and Linux as well.

Not from where I sit. You can still run Linux comfortably on a 486 with
16Mb, which can't be said for any modern version of Winblows.

[snip]
>It is clear that Microsoft OS provides sufficient value added for
>the public to purchase it outright.  It is *not* clear that the Linux
>OS provides sufficient value for an *uneducated* public to acquire
>it for free [*].

Agreed. Linux admin is simply too much of a black art. Not that Winblows
admin is always a cakewalk, but Linux, and Unix in general, really does need
to pull its socks up. A good start might be to begin keeping all those many
configuration files in one central config directory off /etc rather than
scattered all over the disk, but a GUI admin tool is really the only answer.
Even something along the lines of HP's SAM, AIX's SMIT and SCO's SCOADMIN
would be a huge improvement. And it'd be nice if the effort could be
translated to a *standard* tool, so we didn't have to treat each version of
Unix so differently.

Cheers,
Luke



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

From: "Luke Webber" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.ms-windows.nt.advocacy,comp.lang.java.advocacy
Subject: Re: Dvorak calls Microsoft on 'innovation'
Date: Wed, 3 May 2000 16:30:41 +1000

Christopher Browne wrote in message ...
>Indeed.
>
>Although the integration was rather more nearly pioneered by Lotus.
>Anyone remember Symphony?  It was crippled by the DOS 640K limit,
>but combined spreadsheet, word processor, communications/terminal software,
>graphing, and database forms all in one.


Yeah, and M$ responded with Multiplan and a couple of other sort-of
integrated apps which bombed hugely. We have this to thank (or curse) for
the initial development of Windows. Windows wasn't developed just as a GUI,
although that was important. It was also designed to make application
integration easier by allowing multiple apps to be loaded at once. M$ simply
couldn't bear to see somebody beating them out in any corner of the software
market. Some things never change.

>Um.  All of this, including dynamically-linked libraries, was available
>in Multics.
>
>Hint:  Before working on projects that "culminated" in UNIX at Bell Labs,
>Dennis Ritchie worked on Project Mac.  He presumably knew about some of
>the functionality of Multics before starting work on UNIX...


Well there's a very good reason for the similarity of names. I believe the
joke was that they took Multics and cut its balls off. <g>


Luke



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

From: [EMAIL PROTECTED] (abraxas)
Crossposted-To: 
alt.conspiracy,alt.conspiracy.area51,comp.os.ms-windows.advocacy,talk.politics
Subject: Re: Are we equal?
Date: 3 May 2000 06:51:54 GMT

In comp.os.linux.advocacy JEDIDIAH <[EMAIL PROTECTED]> wrote:

>>I understand that.  But I think its far better to talk directly to the 
>>people you have pegged and see what THEY think.  

>       ...and you don't appear to be them.

Glad to see you admit that you do not have me pegged.




=====yttrx


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

From: "Alberto Trillo" <[EMAIL PROTECTED]>
Subject: Clearing things
Date: Wed, 03 May 2000 07:00:07 GMT


   I think most us (GNU/Linux users) do not consider a sin making money
neither with programs nor with computers, though at least I prefer open
source
code if I can use it. The point we did not like (and won't like if Microsoft
finally does not got broken and keeps on its attitude) is companies making
dirty play and hiding API's, perverting protocols, and the so, if the sold a
lot
because they are the best, I guess no one could say nothing; it is the
tricks used
to seem the best what we dislike. Greetings.




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

From: [EMAIL PROTECTED] (abraxas)
Crossposted-To: comp.os.ms-windows.nt.advocacy,comp.lang.java.advocacy
Subject: Re: Is the PC era over?
Date: 3 May 2000 07:06:13 GMT

In comp.os.linux.advocacy Andrew Carpenter <[EMAIL PROTECTED]> wrote:

>> Thats kind of the idea.  Its essentially an X-terminal.

> Even an X-terminal is one step up from this concept, at least in
> functionality stakes. X-terminals do all the windowing (and
> widget-handling?) locally. The SunRay is purely a screenscraper, is it
> not?
>

It does a little more than that, perhaps most notably the ability to 
detatch a running desk-session and re-attach from any other Sun Ray that
can see the server.
  
>> Thats actaully very cheap for X-terminals, which in the past (last time
>> I checked was last year) have gone for between 1000 and 3000 dollars-US.

> I've never worked out why... I don't really know enough about their full
> capabilities. I do know some are capable of actually becoming NCs in
> their own right (applications can be natively compiled and run locally).
>  

I think its mostly because its what the market will bear.  Its something
of a nitche market.

>> X-terminals are not the idea of a 'Network Computer'.  They're the idea
>> of a remote-viewing station.

> That's right... so why should the 'wave of the future' be the equivalent
> of an older technology?

Tesla said the same thing about alot of Edison's work, yet we all still 
use lightbulbs...:)

>> Which is pretty much what an X-terminal does.  The technology for this
>> has been around since the Xwindow system.  Its simple, straightforward,
>> efficient (even moreso now than its ever been) and with recent talk of
>> using an embedded system (linux?) to run an ssh client locally to pipe
>> all your stuff, quite secure.

> An X-terminal can't serve applications or files. Or do you mean the
> X-terminal is the device?

The embedded system would run directly on the diskless X-Terminal client
hardware.  Multiple network protocols can be managed by this kind of
set up, as well as packet encryption mechanism of your choice.  There 
are quite alot of possibilities with an embedded OS like linux (though
admittedly linux's tcp/ip implementation is severely lacking under very
heavy traffic; exactly the sort of thing that this environment 
generates)




=====yttrx

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


** FOR YOUR REFERENCE **

The service address, to which questions about the list itself and requests
to be added to or deleted from it should be directed, is:

    Internet: [EMAIL PROTECTED]

You can send mail to the entire list (and comp.os.linux.advocacy) via:

    Internet: [EMAIL PROTECTED]

Linux may be obtained via one of these FTP sites:
    ftp.funet.fi                                pub/Linux
    tsx-11.mit.edu                              pub/linux
    sunsite.unc.edu                             pub/Linux

End of Linux-Advocacy Digest
******************************

Reply via email to