From: vanev...@blarg.net (Brandon J. Van Every)
Subject: Technologically, how will Linux compete with Windows?
Date: 1996/01/25
Message-ID: <4e96vn$hii@animal.blarg.net>
X-Deja-AN: 137066725
organization: Blarg! Online Services   206/441-9109
newsgroups: comp.os.linux.development.system

Just a word of warning for all you posters hunting for flame-bait.
This is a serious post.  Please answer seriously.

I've been a dedicated Linux hacker for 3 years.  I recently moved to
Seattle, under the shadow of Microsoft.  Because I'm a software
developer I need to keep abreast of other people's products, so I was
forced to install Windows '95 on the teeny 60 MB partition that I have
historically reserved for Windoze 3.1 apps.

First off, it doubled my hard disk space for me...

Then it allowed me to receive faxes.  Easily.  No mgetty and efax
setup hassles - which I must admit, I simply couldn't manage to get
working properly, because I didn't have 3 solid days to play with it
like I used to.  What's more, I could have had the whole Microsoft
Exchange thing integrated into the rest of what I was doing, if I had
wanted.  All right out of the box.

Then I started talking to people on vworlds-biz about freely available
3d graphics libraries and so forth.  I was pointed to Intel's 3DR
technology, which is free, although support for it is being dropped.
Microsoft itself has a whole buttload of 3d technology up its sleeve
right now.  None of it is quite free yet, but I hear the barbarians
crashing at the gate.  There is 3d hardware on the way as well.  In
Christmas 1996, vendors may actually sell it cheap enough that
consumers buy it.

Then I read Microsoft's web pages, and caught up on their corporate
outlook.  Bill Gates has said "we're hard core about the Internet."
They view it as the most strategic business opportunity since the PC
itself.  And they're doing a lot of R&D that's going to weigh very
heavily on what the Internet looks like in the future.  Such as
banking transaction software, advanced web servers, and the ActiveVRML
proposal.  How can Linux hope to compete with any of that?  Sun might
be able to compete, but what about Linux?

Now I am looking for some HTML editing tools, so I can put my resume
online.  I want X11 stuff, of course.  But HTML has changed radically
since I looked at it last May.  Everything is Netscape, everything is
glitzy, with more to come.  My hotlist links to the freeware tools I
used to employ - perfectly good tools at the time - are dead.  Because
it was a familar name, I looked at SoftQuad's HoTMetaL home page.
They have a new free 2.0 version for Windows, a new 2.0 for the Mac,
and UNIX is due sometime this quarter.  Ok, UNIX is 3rd string, that's
really nothing new in the apps realm.

But what does all this say about Linux's ability to "keep up?"

Three years ago when I first started Linux, all other operating
systems basically "sucked."  Nowadays that is no longer the case.
Windows '95, whatever limitations it may have, does not "suck."  There
is nothing about Win 95 that won't get solved by some kind of service
pack or upgrade sometime in late 1996.  And there are lots of "snazzy"
things happening with it, that are going to sell very well in the new
Internet-driven marketplace.

Linux was a success because it made good basic use of system
resources, in a time when no commercial OS really did it "right."  Now
the commercial OS's almost do it "right," and at the same time, the
technological demands for an OS are being taken to a new level.  I see
Linux falling behind technologically.  I don't see where the manpower
to fill this shortfall is going to come from.

Now personally, I have a piece of the puzzle.  I have a C++ 3d
graphics library that's nearing it's first release.  It could serve as
the basis of a freeware project.  Actually it's 100% portable, so it
could serve as the basis for a multi-platform distributed VR freeware
project.  (That was always the original intention.)  But I wonder at
putting forth additional effort on a "sinking ship."  Why not cave in
and develop it primarily under Windows 95?  Worse: since I live in
Seattle, why not cave in and go work for Microsoft's R&D dept.?  Even
if I do "stick to my guns" and develop this stuff as freeware, what is
Linux going to provide as far as infrastructure goes, compared to
Windows 95?  What's the incentive?

Linux needs a vision of its future.  It needs to decide on it's niche.
If for no other reason, than that Microsoft has decided _their_ future
(the Internet), and they are going to steamroller that path, you can
count on it.  Don't take my say-so: read what Bill Gates had to say at
the Internet Strategy Workshop that they had on December 7th, 1995.
<http://www.microsoft.com/Internet/DEFAULT.HTM >.  And then what The
Burton Group had to say about that workshop in terms of market impact.
<http://www.microsoft.com/infoserv/burton1.htm>.

Fellow Linuxers, what are we to do about this?

Cheers,
Brandon

From: ptomb...@compass.xcski.com (Paul Tomblin)
Subject: Re: Technologically, how will Linux compete with Windows?
Date: 1996/01/30
Message-ID: <DM0Ir3.84F@canoe.com>#1/1
X-Deja-AN: 137311583
sender: ptomb...@canoe.com (Paul Tomblin)
references: <4e96vn$hii@animal.blarg.net>
organization: Tomblin Computer Consulting, Rochester, New York and Ottawa, Ontario
reply-to: ptomb...@xcski.com (Paul Tomblin)
newsgroups: comp.os.linux.development.system

In a previous article, vanev...@blarg.net (Brandon J. Van Every) said:
[MicroSoft takes over the world]
>Fellow Linuxers, what are we to do about this?

Nothing.  Linux has always been a haven for people who like to tinker with the
OS, or who love free software.  Windows has never been our competition, and
never will be - {Free,Net}BSD is our competition.

Besides, anybody who can't make a decent web page with vi and a beer doesn't
deserve to run Linux.  :-)

(See http://www.servtech.com/public/ptomblin/rfc.html - I did the logo using a
combination of xpaint and idraw and an existing gif file of a Piper Cherokee I
got off the web.  I made the background using xv (emboss mode) on that same
gif file.)

-- 
Paul Tomblin (ptomb...@xcski.com, formerly ptomb...@canoe.com)
<a href="http://www.servtech.com/public/ptomblin/">My home page</a>
"The superior pilot uses his superior judgement to avoid situations in which
he has to demonstrate his superior skill" - anon.

From: jmala...@cc.helsinki.fi (Jussi Lahtinen)
Subject: Re: Technologically, how will Linux compete with Windows?
Date: 1996/01/31
Message-ID: <4endao$arb@myntti.helsinki.fi>#1/1
X-Deja-AN: 137397656
references: <4e96vn$hii@animal.blarg.net>
content-type: text/plain; charset=ISO-8859-1
organization: University of Helsinki
mime-version: 1.0
newsgroups: comp.os.linux.development.system

In <4e96vn$...@animal.blarg.net> vanev...@blarg.net (Brandon J. Van Every) writes:

>Three years ago when I first started Linux, all other operating
>systems basically "sucked."  Nowadays that is no longer the case.
>Windows '95, whatever limitations it may have, does not "suck."  There

It still sucks.  Its security is nonexsistent (eg.  no file permissions
and process isolation.) This is very serious design mistake.  Remote
usage is very limited too. 

>is nothing about Win 95 that won't get solved by some kind of service
>pack or upgrade sometime in late 1996.  And there are lots of "snazzy"

Security is not an add-on feature. 

Jussi Lahtinen

From: conno...@w3.org (Dan Connolly)
Subject: Distributed Authentication 
[was: Technologically, how will Linux compete with Windows?]
Date: 1996/01/31
Message-ID: <yprnlomn66uw.fsf_-_@beach.w3.org>#1/1
X-Deja-AN: 137582562
sender: conno...@beach.w3.org
references: <4e96vn$hii@animal.blarg.net> <4endao$arb@myntti.helsinki.fi>
organization: W3C
newsgroups: comp.os.linux.development.system

In article <4endao$...@myntti.helsinki.fi> jmala...@cc.helsinki.fi 
(Jussi Lahtinen) writes:
 > In <4e96vn$...@animal.blarg.net> vanev...@blarg.net (Brandon J. Van Every) 
writes:
 > 
 > >Three years ago when I first started Linux, all other operating
 > >systems basically "sucked."  Nowadays that is no longer the case.
 > >Windows '95, whatever limitations it may have, does not "suck."  There
 > 
 > It still sucks.  Its security is nonexsistent (eg.  no file permissions
 > and process isolation.) This is very serious design mistake.  Remote
 > usage is very limited too. 
 > 
 > >is nothing about Win 95 that won't get solved by some kind of service
 > >pack or upgrade sometime in late 1996.  And there are lots of "snazzy"
 > 
 > Security is not an add-on feature. 

Unix security sucks too. The concepts of root, user, group, and other
don't scale to the problem of sofware installation and information
management, and electronic commerce on the global Internet.

At lease MS IE gives little warning dialogs "Are you sure you want
to run this? Do you know where it came from?" before it launches trojan
horses on your system. Unfortunately, users are getting a false
sense security from branded icons and known domain names. DNS spoofing,
IP spoofing -- heck: just hack the distribution on the server. Most
information providers don't even give out MD5s that folks can verify.

Anyway...

The day when I can install a new filesystem on my linux box without logging
in as root, I'll be a happy man. The day when I can transfer credentials
from one window to another via drag-and-drop, I'll be even happier.

Heck: do the modern linux installation mechanisms (Redhat, Debian)
event support network installation (i.e. install a package once,
use it on many manchines. But not just NFS mounting /usr. I want
different groups of packages available to different clients, but
store on the same fileserver. I want different configurations on
different clients, but shared binaries. Stuff like that.)

I read the documentation for plan9 and TAOS, and I drool:

http://achille.research.att.com/plan9/faq.html
=================================================
What about security and user authentication?

Plan 9's authentication design is akin to that of MIT's Kerberos. Passwords are
never sent over networks; instead encrypted tickets are obtained from an
authentication server. It doesn't have the concept of `set UID' programs. The
file server doesn't run user programs, and except at its own console, it doesn't
allow access to protected files except by authenticated owners. The concept
of a special `root' user is gone. 
=================================================


http://gams.cam.nist.gov/acm/Abstracts/0734-2071/138874.html

Authentication in distributed systems : theory
                             and practice

     Lampson, Butler, Abadi, Martmn, Burrows, Michael and Wobber, Edward 

                    ACM Transactions on Computer Systems
                      vol.10, no. 4 (Nov. 1992) pp. 265-310. 


                                   Review

                                 S. A. Kurzban 


Maybe I should read up on GNU hurd or Mach3 about distributed authentication...

Dan
-- 
Daniel W. Connolly        "We believe in the interconnectedness of all things"
Research Scientist, MIT/W3C     PGP: EDF8 A8E4 F3BB 0F3C  FD1B 7BE0 716C FF21 
<conno...@w3.org>                  http://www.w3.org/pub/WWW/People/Connolly/

From: jgarzik@ (Jeff Garzik)
Subject: Re: Technologically, how will Linux compete with Windows?
Date: 1996/02/01
Message-ID: <4ep5mm$3su@brickbat.mindspring.com>#1/1
X-Deja-AN: 137394642
references: <4e96vn$hii@animal.blarg.net> <DM0Ir3.84F@canoe.com>
organization: Slack Central
reply-to: jgar...@pobox.com
newsgroups: comp.os.linux.development.system

In article <DM0Ir3....@canoe.com>, Paul Tomblin <ptomb...@xcski.com> wrote:
>(See http://www.servtech.com/public/ptomblin/rfc.html - I did the logo using a
>combination of xpaint and idraw and an existing gif file of a Piper Cherokee I
>got off the web.  I made the background using xv (emboss mode) on that same
>gif file.)

Yes, but people are going to laugh at you if you try to design a snazzy
corporate site with those same tools.  (Unless you are planning to export a
Sparc Photoshop window to your Linux screen; but that's cheating :))

	Jeff

From: p...@soda.CSUA.Berkeley.EDU (Peter Mattis)
Subject: Re: Technologically, how will Linux compete with Windows?
Date: 1996/02/01
Message-ID: <4er040$kbg@soda.CSUA.Berkeley.EDU>#1/1
X-Deja-AN: 137514580
references: <4e96vn$hii@animal.blarg.net> <DM0Ir3.84F@canoe.com> 
<4ep5mm$3su@brickbat.mindspring.com>
organization: Computer Science Undergraduate Association, UC Berkeley
newsgroups: comp.os.linux.development.system

In article <4ep5mm$...@brickbat.mindspring.com>,
Jeff Garzik <jgar...@pobox.com> wrote:
>In article <DM0Ir3....@canoe.com>, Paul Tomblin <ptomb...@xcski.com> wrote:
>>(See http://www.servtech.com/public/ptomblin/rfc.html - I did the logo using a
>>combination of xpaint and idraw and an existing gif file of a Piper Cherokee I
>>got off the web.  I made the background using xv (emboss mode) on that same
>>gif file.)
>
>Yes, but people are going to laugh at you if you try to design a snazzy
>corporate site with those same tools.  (Unless you are planning to export a
>Sparc Photoshop window to your Linux screen; but that's cheating :))

(shameless plug follows)

True, but you could use the gimp (http://www.xcf.berkeley.edu/~gimp) and
create a way cool site with almost as much ease as with photoshop. (Older
versions of photoshop, that is...still haven't caught up to 3.0 yet. :)
BTW, the logo/picture on the gimp homepage url given above was done 
entirely in the gimp.

Peter Mattis