Newsgroups: comp.os.ms-windows.nt.misc
Path: gmd.de!newsserver.jvnc.net!nntpserver.pppl.gov!princeton!udel!gatech!
swrinde!elroy.jpl.nasa.gov!ames!news.Hawaii.Edu!stahlber
From: stahl...@Hawaii.Edu (Christopher Stahlberger)
Subject: Telnet Daemon for NT
Message-ID: <CMDM4E.7tK@news.Hawaii.Edu>
Sender: n...@news.Hawaii.Edu
Organization: University of Hawaii
X-Newsreader: TIN [version 1.2 PL2]
Date: Wed, 9 Mar 1994 02:41:01 GMT
Lines: 11
Hello everyone,
Subject says it all: is there a Telnet
Daemon for Windows NT? I have checked out the rlogin
daemon and it is pretty cool. But, it's not telnet.
I would think Microsoft is working on this one.
Thanks for any help.
Chris Stahlberger
Newsgroups: comp.os.ms-windows.nt.misc
Path: gmd.de!newsserver.jvnc.net!howland.reston.ans.net!europa.eng.gtefsd.com!
MathWorks.Com!noc.near.net!zcias2.ziff.com!linux.pcweek.ziff.com !esullivan
From: esulli...@linux.pcweek.ziff.com ()
Subject: Re: Telnet Daemon for NT
Message-ID: <CMn4J5.2BH@zcias2.ziff.com>
Sender: n...@zcias2.ziff.com (USENET News System)
Organization: PC Week Labs, Medford, MA.
X-Newsreader: TIN [version 1.2 PL2]
References: <CMDM4E.7tK@news.Hawaii.Edu>
Date: Mon, 14 Mar 1994 05:57:05 GMT
Lines: 26
Christopher Stahlberger (stahl...@Hawaii.Edu) wrote:
[snip]
: daemon and it is pretty cool. But, it's not telnet.
: I would think Microsoft is working on this one.
: Thanks for any help.
: Chris Stahlberger
I asked Microsoft specifically recently if there would be a telnet
daemon (or service, in NT-speak) in the next version (Daytona) and was
told no. They did say someone was working on something like this, but
I wouldn't hold my breath. I think Microsoft isn't planning a telnet
daemon because they don't consider NT a multi-user system, at least in
the Unix sense (maybe in the NetWare sense, though).
--
------------------------------------------------------------------
>Eamonn Sullivan | esulli...@pcweek.ziff.com<
>PC Week Labs | phone: 617-393-3841 <
------------------------------------------------------------------
If you don't have a nasty obituary you probably didn't matter.
-- Freeman Dyson
------------------------------------------------------------------
Newsgroups: comp.os.ms-windows.nt.misc,comp.os.ms-windows.programmer.win32
Path: gmd.de!xlink.net!howland.reston.ans.net!usc!elroy.jpl.nasa.gov!swrinde!
menudo.uh.edu!nuchat!texhrc!ldm
From: l...@texhrc.uucp (Lyle Meier)
Subject: Re: Telnet Daemon for NT
Message-ID: <1994Mar20.063233.17743@texhrc.uucp>
Organization: Texaco EPTD
References: <CMn4J5.2BH@zcias2.ziff.com> <17664@blue.cis.pitt.edu>
Date: Sun, 20 Mar 1994 06:32:33 GMT
Lines: 30
A question: Given that one can monitor a remote nt system adjust its
registry, and run rpc based apps, what specifically do you want to login to
do. Recall that graphics apps will not remote out of the box, although I
believe that with sufficent effort it could be done. (Apps use local
procedure calls to go to the CSRSS that drives the display so if one could
substitute rpc for lpc it could be done, ... sounds like a 3rd party opp).
There do exist several "free ware" rlogin emulators that essentially present
one the image of a console window check ftp.cica.indiana.edu for these.
On unix as I recall one would login to monitor performance, which can be
done with the performance monitor, create users etc which can be done, but
more likley would be running in an NTAS domain where this is done once for
the entire domain, shutdown the system, which cant be done out of the box,
but I would suspect could be done without a lot of difficutly using rpc.
I believe that there is a paradigm shift that needs to occure here for
users,
the idea of command line interfaces is to fade away over time in favor of
gui based interfaces. If this is good or bad is subject to debate which I
don't propose to engage in now. Clearly the market wants this, and wants
simpler system admin than unix.
If one means to login to another nt system to develop code, while it might
be possible to run the tools that underlye visual C++ one could not run the
Integrated development environment or the App studio remotly as they are
GUI based.
I originally thought remote login would be useful, but after looking at the
admin tools in nt and the resource kit, the question became what could I do
that cant be done another way? I am curious to know what those who have
used nt believe to be the need for remote login. Thank you.
Path: gmd.de!xlink.net!news.ppp.de!lutzifer!news.rrz.uni-hamburg.de!news.dkrz.de!
news.dfn.de!zeus.rbi.informatik.uni-frankfurt.de!terra.wiwi.uni-frankfurt.de!
news.th-darmstadt.de!zib-berlin.de!netmbx.de!Germany.EU.net!EU.net!
howland.reston.ans.net!gatech!swrinde!sgiblab!barrnet.net!nntp.crl.com!crl.crl.com!
not-for-mail
From: dwil...@crl.com (Dick Wilmot)
Newsgroups: comp.os.ms-windows.nt.misc,comp.os.ms-windows.programmer.win32
Subject: Re: Telnet Daemon for NT
Date: 20 Mar 1994 11:32:33 -0800
Organization: CRL Dialup Internet Access
Lines: 92
Message-ID: <2mi8ch$e05@crl.crl.com>
References: <CMn4J5.2BH@zcias2.ziff.com> <17664@blue.cis.pitt.edu>
<1994Mar20.063233.17743@texhrc.uucp>
NNTP-Posting-Host: crl.crl.com
l...@texhrc.uucp (Lyle Meier) writes:
>I believe that there is a paradigm shift that needs to occure here for
>users, ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
The paradigm shift needs to occur in Redmond, WA
This reminds me of another large computer company in Armonk, NY that thought
it knew better than its customers what they needed in place of what they
wanted - though not out of business yet IBM found out where the paradigm gets
stuck when you try to dictate to customers.
>the idea of command line interfaces is to fade away over time in favor of
>gui based interfaces. If this is good or bad is subject to debate which I
>don't propose to engage in now. Clearly the market wants this, and wants
>simpler system admin than unix.
The market might well want easier system administration but they will get my
CMD away from me only when they prohibit programming. The thing I HATED about
MacInstops was that they were EXPERT HOSTILE and, especially in their
early days, required me to mouse around to merely scroll some text. One of
the things I like about NT is that I can stop launching applications in
favor of STARTING them from the command line. I can type a hell of a lot
faster than I can mouse. Mousing is fine the first few times I use an app
but when I use something 20 times per day and it involves lots of keying
such as in word processing, code development, spread sheeting, starting
applications. And before anyone tries to replace my keyboard with a
microphone or my display screen with a loudspeaker consider that I can
type as fast as I can talk and can read much faster than intelligible
speech. Look out for bandwidths. Anyone who gets paid essentially by the
keystroke as I do is looking for bandwidth. If you force me to use a GUI
then it should have lots of accelerators so I don't have to lift my
gingers from the keyboard for anything.
>If one means to login to another nt system to develop code, while it might
>be possible to run the tools that underlye visual C++ one could not run the
>Integrated development environment or the App studio remotly as they are
>GUI based.
So why can't the GUI be remote. A colleague might want to do NT
development on my machine remotely from her Windows for Workgroups machine. I
might want to do XWindows development on a system down the hall. I am not
a big Unix fan but the disappearance of unix is a fantasy that is only
believed in Washington State - comes from the constant rain or maybe it's
the Tumwater. Lack of remote windows is a deficiency of Windows and NT
not an advantage.
Even better than remote windows would be cooperative windows where I can
show you something I am working on by sharing the window with you and we
can share revisions in real time. Today I find myself having to walk
around cubicles to debug code or editorial copy.
Maybe I want to develop NT programs for the Alpha machine down the hall.
The MS model would require me to buy a duplicate machine for my office or
walk down the hall and chase the current occupant away. I am about to
ship a Postscript file to a friend in Japan to get around the fact that I
can't let him open a window from his MacIntrash machine (even if it were
in the same room).
>I originally thought remote login would be useful, but after looking at the
>admin tools in nt and the resource kit, the question became what could I do
>that cant be done another way? I am curious to know what those who have
>used nt believe to be the need for remote login. Thank you.
Maybe it's a matter of more imagination and not liking procrustean solutions.
If MS will supply the highway we can show them how you use it (see above
and consider how my friend with the Atari can share her latest dazzling
video without remote windows). Think about Word for Workgroups or sharing
1-2-3 SESSIONS (not files) among accountants, auditors and financial
planners. The easiest way of doing these things is NOT OLE 2.0 but rather
remote windows.
I wouldn't be surprised to find third parties at work on this. Windows
and NT are mostly message-based systems at the API level except for some
things like SetWindowPos, MoveWindow, etc., so it might not be so
difficult to hook the message queue messages and duplicate them across
networks.
There is, however, a big missing piece: remote windows should not mean
that a remote user should have access to all "my" resources. There seems
to be no API for logging on a second, third, ... user under their IDs. We
need some form of setuserid (SU) that allows processes to be created for
remote users in parallel with the user at the desk. If Microsoft provided
that API then third parties could more quickly and safely provide the
remote windows we need.
So far MS reaction is to keep repeating their mantra: "One user one machine."
whilst ignoring the uses for remote/cooperative windows. If they won't
help build the highway MS could get run over.
--
Dick Wilmot
Editor, Independent RAID Report
(510) 938-7425
Path: gmd.de!newsserver.jvnc.net!nntpserver.pppl.gov!princeton!udel!MathWorks.Com!
europa.eng.gtefsd.com!howland.reston.ans.net!news.intercon.com!psinntp!psinntp!
psinntp!heimdall!axpo10.sdrc.com!crbalsn
From: crba...@axpo10.sdrc.com (Jim Balson)
Newsgroups: comp.os.ms-windows.nt.misc,comp.os.ms-windows.programmer.win32
Subject: Re: Telnet Daemon for NT
Message-ID: <7482@heimdall.sdrc.com>
Date: 23 Mar 94 16:42:19 GMT
References: <CMn4J5.2BH@zcias2.ziff.com> <17664@blue.cis.pitt.edu>,
<1994Mar20.063233.17743@texhrc.uucp> <2mj3hl$kfn@oac4.hsc.uth.tmc.edu>
Sender: n...@heimdall.sdrc.com
Reply-To: crba...@axpo10.sdrc.com (Jim Balson)
Organization: OrgFreeware
Lines: 98
>>If one means to login to another nt system to develop code, while it might
>>be possible to run the tools that underlye visual C++ one could not run the
>>Integrated development environment or the App studio remotly as they are
>>GUI based.
>
>So why can't the GUI be remote. A colleague might want to do NT
>development on my machine remotely from her Windows for Workgroups machine. I
>might want to do XWindows development on a system down the hall. I am not
>a big Unix fan but the disappearance of unix is a fantasy that is only
>believed in Washington State - comes from the constant rain or maybe it's
>the Tumwater. Lack of remote windows is a deficiency of Windows and NT
>not an advantage.
Absolutely correct. And don't look for this to change anytime soon.
Probably not within this decade anyway. I have read where Gates and Co. don't
beleive in the the philosophy of the X was designed where one could do
these types of things remotly. A major disadvantage.
>Even better than remote windows would be cooperative windows where I can
>show you something I am working on by sharing the window with you and we
>can share revisions in real time. Today I find myself having to walk
>around cubicles to debug code or editorial copy.
>
>Maybe I want to develop NT programs for the Alpha machine down the hall.
>The MS model would require me to buy a duplicate machine for my office or
>walk down the hall and chase the current occupant away. I am about to
>ship a Postscript file to a friend in Japan to get around the fact that I
>can't let him open a window from his MacIntrash machine (even if it were
>in the same room).
>
>>I originally thought remote login would be useful, but after looking at the
>>admin tools in nt and the resource kit, the question became what could I do
>>that cant be done another way? I am curious to know what those who have
>>used nt believe to be the need for remote login. Thank you.
>
Perhaps to do sys admin without having to physically walk over to the
machine? Perhaps to run some application that isn't available over the net?
In the Unix/X world, I frequently find myself logging into other machines,
running application over the network and displaying the output on my screen
where ever I happen to be sitting. Very nice feature.
>I wouldn't be surprised to find third parties at work on this. Windows
>and NT are mostly message-based systems at the API level except for some
>things like SetWindowPos, MoveWindow, etc., so it might not be so
>difficult to hook the message queue messages and duplicate them across
>networks.
>
>There is, however, a big missing piece: remote windows should not mean
>that a remote user should have access to all "my" resources. There seems
>to be no API for logging on a second, third, ... user under their IDs. We
>need some form of setuserid (SU) that allows processes to be created for
>remote users in parallel with the user at the desk. If Microsoft provided
>that API then third parties could more quickly and safely provide the
>remote windows we need.
But wouldn't that be a hack? I don't understand the inner workings
of NT (I'm a Unix/X person) all that well to know for sure. And how do you
envision this "remote windows" to work? Do you suggest changing the Windows
API?
In X, you specify an environment variable where you want the window'ed
output to appear (DISPLAY=machine:0, XOpenDisplay("machine:0", ...). How
could something like that be done with MS windows, which generally is
ignorant of networks?
>So far MS reaction is to keep repeating their mantra: "One user one machine."
>whilst ignoring the uses for remote/cooperative windows. If they won't
>help build the highway MS could get run over.
Yep, it's quite a shame to have a 200 MIP machine on my desk where
only I can use it. Certainly with that much horsepower, you could put a
half dozen people on there without feeling much side effects. Thats where
the Unix/X machines really shines. Unix being multiuser and all, I could
connect as many X Terminals physically possible and use 1 machine to
support a small (5-6) development group. Same goes for OS/2.
>
>--
>Dick Wilmot
>Editor, Independent RAID Report
>(510) 938-7425
Jim
jim.bal...@sdrc.com
Newsgroups: comp.os.ms-windows.nt.misc,comp.os.ms-windows.programmer.win32
Path: gmd.de!xlink.net!howland.reston.ans.net!europa.eng.gtefsd.com!MathWorks.Com!
noc.near.net!zcias2.ziff.com!esullivan
From: esulli...@linux.pcweek.ziff.com
Subject: Re: Telnet Daemon for NT
In-Reply-To: crbalsn@axpo10.sdrc.com's message of 23 Mar 94 16:42:19 GMT
Message-ID: <ESULLIVAN.94Mar24024201@linux.pcweek.ziff.com>
Sender: n...@zcias2.ziff.com (USENET News System)
Organization: PC Week Labs
References: <CMn4J5.2BH@zcias2.ziff.com>
<17664@blue.cis.pitt.edu>,<1994Mar20.063233.17743@texhrc.uucp>
<2mj3hl$kfn@oac4.hsc.uth.tmc.edu> <7482@heimdall.sdrc.com>
Date: Thu, 24 Mar 1994 07:42:01 GMT
Lines: 23
In article <7...@heimdall.sdrc.com> crba...@axpo10.sdrc.com (Jim Balson) writes:
> Yep, it's quite a shame to have a 200 MIP machine on my desk where
> only I can use it. Certainly with that much horsepower, you could put a
> half dozen people on there without feeling much side effects. Thats where
> the Unix/X machines really shines. Unix being multiuser and all, I could
> connect as many X Terminals physically possible and use 1 machine to
> support a small (5-6) development group. Same goes for OS/2.
> ^^^^^^^^^^^^^^^^^^^
I was following you just fine up to this point. My understanding of
OS/2 (which is fairly good) is that it has the same problem as NT
does: it isn't multiuser. So several people can work on one OS/2
system (using that what's-its-name product (cytrix?)), but they all
have the same permissions. Is this not so?
--
---------------------------------------------------------
Eamonn Sullivan | esulli...@pcweek.ziff.com
PC Week Labs | phone: 617-393-3841
---------------------------------------------------------
Newsgroups: comp.os.ms-windows.nt.misc,comp.os.ms-windows.programmer.win32
Path: gmd.de!newsserver.jvnc.net!nntpserver.pppl.gov!princeton!att-in!news.bu.edu!
olivea!charnel!yeshua.marcam.com!MathWorks.Com!europa.eng.gtefsd.com!
howland.reston.ans.net!agate!tfs.com!usenet
From: mmo...@tfs.com
Subject: Re: Telnet Daemon for NT
Message-ID: <Cn6o94.JCz@tfs.com>
Sender: use...@tfs.com
Organization: TRW Financial Services, Inc.
X-Newsreader: <WinQVT/Net v3.9>
Date: Thu, 24 Mar 1994 19:17:28 GMT
Lines: 24
In article <ESULLIVAN.94Mar24024...@linux.pcweek.ziff.com>
esulli...@linux.pcweek.ziff.com writes:
>I was following you just fine up to this point. My understanding of
>OS/2 (which is fairly good) is that it has the same problem as NT
>does: it isn't multiuser. So several people can work on one OS/2
>system (using that what's-its-name product (cytrix?)), but they all
>have the same permissions. Is this not so?
Software Innovations Inc. has a telnetd for NT now that allows as
many users as you have system resources for. You can run any
application that can redirect stdio. That means you can run just
about anything that will runs with a command line interface.
They also have rlogind rexecd rshd and rcp. All of this and a bunch
more goodies for $150.
Software Innovations Inc. (800) 946-6688
Disclaimer: I have no association with SII whatsoever, except that I've
read their product literature.
Newsgroups: comp.os.ms-windows.nt.misc,comp.os.ms-windows.programmer.win32
Path: gmd.de!xlink.net!howland.reston.ans.net!agate!tfs.com!usenet
From: mmo...@tfs.com
Subject: Re: Telnet Daemon for NT
Message-ID: <Cn8nzJ.BHI@tfs.com>
Sender: use...@tfs.com
Organization: TRW Financial Services, Inc.
X-Newsreader: <WinQVT/Net v3.9>
Date: Fri, 25 Mar 1994 21:06:55 GMT
Lines: 33
In article <ESULLIVAN.94Mar25123...@linux.pcweek.ziff.com>
esulli...@linux.pcweek.ziff.com writes:
>In article <Cn6o94....@tfs.com> mmo...@tfs.com writes:
>
>> Software Innovations Inc. has a telnetd for NT now that allows as
>> many users as you have system resources for. You can run any
>> application that can redirect stdio. That means you can run just
>> about anything that will runs with a command line interface.
>> They also have rlogind rexecd rshd and rcp. All of this and a bunch
>> more goodies for $150.
>>
>> Software Innovations Inc. (800) 946-6688
>>
>> Disclaimer: I have no association with SII whatsoever, except that I've
>> read their product literature.
>
>This sounds no different than nrlogin that's available for free. Does
>it allow multiple users to be logged in (with different access
>permissions)? I don't think that's possible on NT.
Wrong! I just talk to Software Innovations (very nice people BTW) and
they said each user has their own security token, separate permissions
on files, directories, whatever.
They also have as separate products:
UUCP
NNTP server and News reader
lpr lpd
DNS
Before long various third parties will have supplied us with all the
nice things Microsoft left out.
Michael M.
Newsgroups: comp.os.ms-windows.nt.misc,comp.os.ms-windows.programmer.win32
Path: gmd.de!newsserver.jvnc.net!netnews.upenn.edu!news.amherst.edu!
news.mtholyoke.edu!news.byu.edu!news.kei.com!MathWorks.Com!europa.eng.gtefsd.com!
howland.reston.ans.net!pipex!uunet!epimbe!vlcek
From: vl...@epimbe.com (James Vlcek)
Subject: Re: Telnet Daemon for NT
Message-ID: <CnC9Ky.5qJ@epimbe.com>
Sender: vl...@epimbe.com (Jim Vlcek)
Organization: EPI Software Systems
References: <Cn8nzJ.BHI@tfs.com>
Date: Sun, 27 Mar 1994 19:46:10 GMT
Lines: 25
In article <Cn8nzJ....@tfs.com> mmo...@tfs.com writes:
>I just talk to Software Innovations (very nice people BTW) and
>they said each user has their own security token, separate permissions
>on files, directories, whatever.
How did they manage this? My (perhaps incorrect) impression was that
Microsoft had not documented the security/authentication APIs
sufficiently for a third party to accomplish this.
If they _have_ accomplished this, is there any reason why a UNIX-like
`su' program could not be written? In fact, what then distinguishes
NT from a true multi-user OS?
>Before long various third parties will have supplied us with all the
>nice things Microsoft left out.
Be forewarned, however: UNIX left out a lot of "nice things" too
(although not telnetd :-). The end result was we got a plethora of
solutions, all achieving the same end, and all differing in
functionality and user interface to a variety of degrees. The
confusion and headaches this resulted in is one of the reasons NT even
exists today.
Jim Vlcek
vl...@epimbe.com
Path: gmd.de!xlink.net!howland.reston.ans.net!pipex!sunic!psinntp!psinntp!congrunt!artk
From: a...@Congruent.COM (Arthur Kreitman)
Newsgroups: comp.os.ms-windows.nt.misc
Subject: Re: Telnet Daemon for NT
Message-ID: <ARTK.94Mar28212010@cc-color1.Congruent.COM>
Date: 29 Mar 94 02:20:10 GMT
References: <CMn4J5.2BH@zcias2.ziff.com> <17664@blue.cis.pitt.edu>
<1994Mar20.063233.17743@texhrc.uucp> <2mj3hl$kfn@oac4.hsc.uth.tmc.edu>
<2mkkno$ng9@strauss.udel.edu> <hamilton.764351154@BIX.com>
<ARTK.94Mar25090203@cc-color1.Congruent.COM> <1994Mar25.150105.1423
Sender: n...@Congruent.COM
Organization: Congruent Corporation; New York, NY
Lines: 28
In-reply-to: hamilton@BIX.com's message of 27 Mar 94 14:35:59 GMT
brian%cons.cs.utah.edu@cs.utah.edu (Brian Sturgill) writes:
>Ataman Software, L.C. (a commercial/shareware software firm belonging
>to my wife and I) will soon be releasing an rlogind and telnetd that can
>handle all that "fancy" stuff. An "rexecd" should be released for beta
>this weekend.
>--
>C. Brian Sturgill
>University of Utah Windows family OS info -- ftp to easy.cs.utah.edu.
>Center for Software
>Science brian@cs.utah.edu
Now that's what every buyer want's to hear, that a critical product
for their firm comes from a company that consists of a student at a
grade b CS school operating from a kitchen table. And the dog does
tech support.
--
----
Art Kreitman Congruent Corporation
a...@congruent.com 110 Greene Street
212-431-5100 New York, New York 10012
fax 219-1532
Newsgroups: comp.os.ms-windows.nt.misc
Path: bga.com!news.sprintlink.net!qns1.qns.com!constellation!paladin.american.edu!
howland.reston.ans.net!math.ohio-state.edu!magnus.acs.ohio-state.edu!csn!att-in!
att-out!pacbell.com!uop!csus.edu!netcom.com!jeremy
From: jer...@netcom.com (Jeremy Allison)
Subject: Re: Telnet Daemon for NT
Message-ID: <jeremyCnG0t7.Kxo@netcom.com>
Organization: NETCOM On-line Communication Services (408 241-9760 guest)
References: <CMn4J5.2BH@zcias2.ziff.com> <17664@blue.cis.pitt.edu>
<1994Mar20.063233.17743@texhrc.uucp> <2mj3hl$kfn@oac4.hsc.uth.tmc.edu>
<2mkkno$ng9@strauss.udel.edu> <hamilton.764351154@BIX.com>
<ARTK.94Mar25090203@cc-color1.Congruent.COM>
Date: Tue, 29 Mar 1994 20:27:06 GMT
Lines: 24
a...@Congruent.COM (Arthur Kreitman) writes:
> Login is easy.
>--
Dear Mr Kreitman,
If "Login is easy" could you please post some
pseudocode showing everyone the Win32 API calls/ registry
manipulation you need to do in order to change the security
token of a process from administrator to another user id, given
that users name and plaintext password.
I'm sure *many* people here would be extremely interested in
how it was done, as much pd software depends on it.
(I need a working NT inetd for instance).
Thanks in advance,
Jeremy Allison,
jer...@netcom.com
Path: gmd.de!newsserver.jvnc.net!netnews.upenn.edu!msuinfo!uwm.edu!
fnnews.fnal.gov!d0xs2.fnal.gov!jpb
From: j...@d0xs2.fnal.gov (John Borders)
Newsgroups: comp.os.ms-windows.nt.misc,comp.os.ms-windows.programmer.win32
Subject: Re: Telnet Daemon for NT
Date: 1 Apr 1994 19:17:19 GMT
Organization: Fermi National Lab/D-Zero Experiment
Lines: 47
Distribution: world
Message-ID: <2nhrvv$f37@fnnews.fnal.gov>
References: <Cn8nzJ.BHI@tfs.com> <CnC9Ky.5qJ@epimbe.com>
Reply-To: j...@d0xs2.fnal.gov (John Borders)
NNTP-Posting-Host: d0xs2.fnal.gov
X-Newsreader: mxrn 6.18-4
:In article <CnC9Ky....@epimbe.com>, vl...@epimbe.com (James Vlcek) writes:
:In article <Cn8nzJ....@tfs.com> mmo...@tfs.com writes:
:>Before long various third parties will have supplied us with all the
:>nice things Microsoft left out.
:
:Be forewarned, however: UNIX left out a lot of "nice things" too
:(although not telnetd :-). The end result was we got a plethora of
:solutions, all achieving the same end, and all differing in
:functionality and user interface to a variety of degrees. The
:confusion and headaches this resulted in is one of the reasons NT even
:exists today.
This is a VERY good point, that hadn't occurred to me. I hope Microsoft
takes this under consideration. If too many people use an aspect of NT
which has too many different third-party implementations, we are back to
the point one has to be an expert in all the obscure products to really
be able to administrate an operating system. It also makes it very
unattractive when we have to call different third-party suppliers for
every problem with a different implementation. We would rather PAY one
dealer for support (MS, in this case), than rely on support from many other
parties.
One of the attractions of NT for us is its simplicity. However, we NEED
telnetd, in order to utilize all the other workstations and xterminals we
have presently (literally hundreds) for the next few years until they can all
be replaced with NT-capable stations. Of course, it will take even longer
for all the users to get used to programming in an RPC-centric manner (I'm
not even addressing the issue of whether this is indeed superior).
Granted, I am in the academic world (high energy physics), and perhaps
Microsoft does not consider us their prime market (we may even qualify as a
negligible market), but they should consider the value the commercial and
industrial fields place upon academic experts. If they are trying to take over
the market that mainframes, then workstations, used to dominate, they
shouldn't take the attitude that the very limited personnel at MS know best.
I haven't heard anyone from anywhere supporting Microsoft's contention that
multi-user capabilities are obsolete, and I hope they are not so arrogant to
simply assume they know better than all the other experts in the world
(I'm not qualifying myself as an expert, but I believe many of the people
who have posted in this group probably do so qualify).
:Jim Vlcek
:vl...@epimbe.com
John Borders
j...@fnald0.fnal.gov
Path: gmd.de!newsserver.jvnc.net!netnews.upenn.edu!msuinfo!uwm.edu!
vixen.cso.uiuc.edu!howland.reston.ans.net!europa.eng.gtefsd.com!news.ans.net!
malgudi.oar.net!witch!ankh!janb
Newsgroups: comp.os.ms-windows.nt.misc,comp.os.ms-windows.programmer.win32
Message-ID: <296@ankh.win.net>
References: <Cn8nzJ.BHI@tfs.com> <CnC9Ky.5qJ@epimbe.com><2nhrvv$f37@fnnews.fnal.gov>
Reply-To: j...@ankh.win.net (Jan Bottorff)
From: j...@ankh.win.net (Jan Bottorff)
Date: Sat, 02 Apr 1994 21:33:53 GMT
Subject: Re: Telnet Daemon for NT
Lines: 69
In article <2nhrvv$...@fnnews.fnal.gov>, John Borders (j...@d0xs2.fnal.gov) writes:
>
> ...
>I haven't heard anyone from anywhere supporting Microsoft's contention that
>multi-user capabilities are obsolete, and I hope they are not so arrogant to
>simply assume they know better than all the other experts in the world
>(I'm not qualifying myself as an expert, but I believe many of the people
>who have posted in this group probably do so qualify).
>
I don't think MSFT is saying that multi-user computers are obsolete.
For example, it seems like they don't expect 16 pentium processor
machines running NT (from AT&T or Sequent) to be for a single
user. I think their stand is more along the lines of "sharing a
computer degrades the user-interface", which I would have to agree
with. It just seems like MSFT has a different opinion of WHERE a
machine boundry should be:
-------------
| low level |
| graphics |
------------- <- X-Windows splits here
| user |
| interface |
------------- <- MSFT splits here
| data |
| model |
-------------
I personally believe it's NOT acceptable for the user-interface to
stop responding because your network happens to be busy or a remote
computer just went away. It also seems like the bandwidth
requirements (and REAL-TIME requirements) between the low-level
graphics/event input is rather higher than the bandwidth between the
user interface and the data model. I do believe it would be possible
to utilize the wasted CPU cycles on the local machine a bit
better. This seems more like a problem of dynamically scheduling
computational "packets" across a network better. It doesn't seem
like Unix/X-Windows has this especially together. One of the
problems seem like the popular programming languages have not made
it easy to distribute "parts" of the program across the network.
For example, it seems reasonable to say something like this (in
Smalltalk):
theImage1 := [Fractal computeMandel1] usingNetworkResources.
theImage2 := [Fractal computeMandel2] usingNetworkResources.
"do some other things in preparation of displaying"
myWindow graphicsContext display:theImage1;
display:theImage2.
This would initiate the computation of two fractals across your
network storing a proxy for the results. We would then display the
results overlapping the display time with the calculation time.
The display code could execute until it REALLY needs the data,
blocking this process if the data wasn't ready yet or not blocking
if it was, or displaying what was available and telling the proxy
to notify the display code when the data DID become available. I
can think of LOTS of things that could be broken down like this if
it were easy (and fast). Just RUNNING the program with a remote
input/display connection doesn't solve the distributed computing
problem.
- Jan
___________________________________________________________________
Jan Bottorff Internet:j...@netcom.com
Paradigm Matrix CIS: 74775,546
15561 Lori Anne Lane voice (408) 272-8523
San Jose, CA 95127-2606 fax (408) 272-2371
Path: gmd.de!nntp.gmd.de!xlink.net!news.dfn.de!darwin.sura.net!
howland.reston.ans.net!europa.eng.gtefsd.com!MathWorks.Com!mvb.saic.com!
news.cerf.net!maxwell.expersoft.com!usenet
From: d...@expersoft.com
Newsgroups: comp.os.ms-windows.programmer.win32,comp.os.ms-windows.nt.misc
Subject: Re: Telnet Daemon for NT
Date: Mon, 11 Apr 94 15:21:22
Organization: Expersoft
Lines: 78
Message-ID: <2ocm1r$c0v@chomolangma.expersoft.com>
References: <Cn8nzJ.BHI@tfs.com>
<CnC9Ky.5qJ@epimbe.com><2nhrvv$f37@fnnews.fnal.gov> <296@ankh.win.net>
NNTP-Posting-Host: zen.expersoft.com
Mime-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
>
> I don't think MSFT is saying that multi-user computers are obsolete.
> For example, it seems like they don't expect 16 pentium processor
> machines running NT (from AT&T or Sequent) to be for a single
> user. I think their stand is more along the lines of "sharing a
> computer degrades the user-interface"...
>
> I personally believe it's NOT acceptable for the user-interface to
> stop responding because your network happens to be busy or a remote
> computer just went away. It also seems like the bandwidth
> requirements (and REAL-TIME requirements) between the low-level
> graphics/event input is rather higher than the bandwidth between the
> user interface and the data model. I do believe it would be possible
> to utilize the wasted CPU cycles on the local machine a bit
> better. This seems more like a problem of dynamically scheduling
> computational "packets" across a network better. It doesn't seem
> like Unix/X-Windows has this especially together.
If you talk to Unix developers about NT, one of the first questions
90% percents of them ask is if NT supports remote logins. Actually,
there is nothing wrong with it - it's just a Unix Culture. The idea
was like you buy one huge and powerful server and bunch of dumb and
cheap X-terminals which could display stuff but could not think.
This so-to-say Mainframism is a nice theory which works and at the
same time doesn't. I've been working for Unix companies for 2 years
(doing Windows development for them) and I did not see a developer
which would have x-term on his desk. Developers usually need Unix
boxes with it's own CPU and stuff. So, x-terms were supposed to be
used by users. But how many Unix users did you have a chance to see ?
PCs are such that they have their own brains and there is no big
need for them to borrow CPU cycles from anyone else. I think that
remote-logins and X-Terminalism is a cool stuff but it's more like
it just would be nice to have them (not mentioning that they have
their own problems).
> One of the
> problems seem like the popular programming languages have not made
> it easy to distribute "parts" of the program across the network.
> For example, it seems reasonable to say something like this (in
> Smalltalk):
>
> theImage1 := [Fractal computeMandel1] usingNetworkResources.
> theImage2 := [Fractal computeMandel2] usingNetworkResources.
>
> "do some other things in preparation of displaying"
>
> myWindow graphicsContext display:theImage1;
> display:theImage2.
>
> This would initiate the computation of two fractals across your
> network storing a proxy for the results...
Sounds like you need XShell. XShell for Windows is already available (beta).
XShell provides natural extension to C++, so that you could do something
like this:
class Fractal
{
private:
//...
public:
Fractal();
~Fractal();
distributed:
Calculate();
};
Regards,
Daniel Bodrov
Path: gmd.de!nntp.gmd.de!dearn!
xlink.net!news.ppp.de!lutzifer!news.rrz.uni-hamburg.de!news.dkrz.de!
news.dfn.de!darwin.sura.net!howland.reston.ans.net!cs.utexas.edu!uunet!
ncrgw2.ncr.com!ncrhub2!ncr-mpd!Don.Allingham
From: Don.Alling...@FtCollinsCO.NCR.COM (Don Allingham)
Newsgroups: comp.os.ms-windows.programmer.win32,comp.os.ms-windows.nt.misc
Subject: Re: Telnet Daemon for NT
Message-ID: <DON.ALLINGHAM.94Apr12080054@geebee.FtCollinsCO.NCR.COM>
Date: 12 Apr 94 14:00:54 GMT
References: <Cn8nzJ.BHI@tfs.com> <CnC9Ky.5qJ@epimbe.com><2nhrvv$f37@fnnews.fnal.gov>
<296@ankh.win.net> <2ocm1r$c0v@chomolangma.expersoft.com>
Sender: u...@ncr-mpd.FtCollinsCO.NCR.COM
Organization: NCR Microelectronics Products Division
Lines: 53
In-reply-to: dnb@expersoft.com's message of 11 Apr 94 14:21:22 GMT
>>>>> "dnb" == dnb <d...@expersoft.com> writes:
dnb> If you talk to Unix developers about NT, one of the first
dnb> questions 90% percents of them ask is if NT supports remote
dnb> logins. Actually, there is nothing wrong with it - it's just
dnb> a Unix Culture. The idea was like you buy one huge and
dnb> powerful server and bunch of dumb and cheap X-terminals which
dnb> could display stuff but could not think. This so-to-say
dnb> Mainframism is a nice theory which works and at the same time
dnb> doesn't. I've been working for Unix companies for 2 years
dnb> (doing Windows development for them) and I did not see a
dnb> developer which would have x-term on his desk. Developers
dnb> usually need Unix boxes with it's own CPU and stuff. So,
dnb> x-terms were supposed to be used by users. But how many Unix
dnb> users did you have a chance to see ?
dnb> PCs are such that they have their own brains and there is no
dnb> big need for them to borrow CPU cycles from anyone else. I
dnb> think that remote-logins and X-Terminalism is a cool stuff
dnb> but it's more like it just would be nice to have them (not
dnb> mentioning that they have their own problems).
I'm sorry. I cannot let this go. Remote login is necessary for a
large network if for no other reason than for system administration.
We have over 100 (probably closer to 150) Sun and HP workstations in
three buildings. Using remote login, most of the administration can
be done by the administrators in their own office, instead of having
to run around to 100+ offices in three separate buildings.
And as far as my Sparc 10 or HP 735 not having CPU cycles ....
One of the big problems we have that requires remote logins is the
cost of software. In my job of circuit design, we tend to use some
pretty pricey software ($100K+). There are 15 engineers who need
access to this software. We cannot spend $1.5M to give each
engineer a copy, especially when they do not use it all the time.
An everyone standing in line to use one machine's console is not
practical either. That is where remote logins are necessary.
For the most part, most people can do what they need to do on their
own machine. However, this is not always the case.
Please do not make global generalizations on other people's needs
based on your own.
--
Don.Alling...@FtCollins.NCR.com
NCR Microelectronics
AT&T Global Information Solutions
Ft. Collins, CO.
Newsgroups: comp.os.ms-windows.programmer.win32,comp.os.ms-windows.nt.misc
Path: gmd.de!nntp.gmd.de!xlink.net!howland.reston.ans.net!cs.utexas.edu!
swrinde!ihnp4.ucsd.edu!library.ucla.edu!csulb.edu!csus.edu!netcom.com!tsikes
From: tsi...@netcom.com (Terry Sikes)
Subject: Re: Telnet Daemon for NT
Message-ID: <tsikesCoMIB6.C16@netcom.com>
Organization: NETCOM On-line Communication Services (408 241-9760 guest)
References: <Cn8nzJ.BHI@tfs.com> <296@ankh.win.net>
<2ocm1r$c0v@chomolangma.expersoft.com>
<DON.ALLINGHAM.94Apr12080054@geebee.ftcollinsco.ncr.com>
Date: Thu, 21 Apr 1994 19:04:18 GMT
Lines: 21
In article <DON.ALLINGHAM.94Apr12080...@geebee.ftcollinsco.ncr.com>,
Don Allingham <Don.Alling...@FtCollinsCO.NCR.COM> wrote:
[compress]
>One of the big problems we have that requires remote logins is the
>cost of software. In my job of circuit design, we tend to use some
>pretty pricey software ($100K+). There are 15 engineers who need
>access to this software. We cannot spend $1.5M to give each
>engineer a copy, especially when they do not use it all the time.
>An everyone standing in line to use one machine's console is not
>practical either. That is where remote logins are necessary.
PMJI, but why is a license server not an option here? No remote login,
just remote execution...
--
Terry Sikes == tsi...@netcom.com | All I want is a baseline 1/1/1 machine:
Alternate address: tsi...@fatcity.com | 1000 MIPS, 1000 MB RAM, and 1000 GB HD.
Also tlsi...@bix.com, tsi...@aol.com |
All my opinions reflect my views only!| Is that too much to ask?
Path: gmd.de!nntp.gmd.de!xlink.net!howland.reston.ans.net!agate!msuinfo!
harbinger.cc.monash.edu.au!news.uwa.edu.au!info.curtin.edu.au!cc.curtin.edu.au!
zcookbruc
From: zcookb...@cc.curtin.edu.au
Newsgroups: comp.os.ms-windows.programmer.win32,comp.os.ms-windows.nt.misc
Subject: Re: Telnet Daemon for NT
Date: 23 Apr 94 02:53:20 +0900
Organization: Curtin University of Technology
Lines: 32
Message-ID: <1994Apr23.025320.1@cc.curtin.edu.au>
References: <Cn8nzJ.BHI@tfs.com> <296@ankh.win.net>
<2ocm1r$c0v@chomolangma.expersoft.com>
<DON.ALLINGHAM.94Apr12080054@geebee.ftcollinsco.ncr.com> <tsikesCoMIB6.C16@netcom.com>
NNTP-Posting-Host: cc.curtin.edu.au
In article <tsikesCoMIB6....@netcom.com>, tsi...@netcom.com (Terry Sikes) writes:
> In article <DON.ALLINGHAM.94Apr12080...@geebee.ftcollinsco.ncr.com>,
> Don Allingham <Don.Alling...@FtCollinsCO.NCR.COM> wrote:
>
> [compress]
>
>>One of the big problems we have that requires remote logins is the
>>cost of software. In my job of circuit design, we tend to use some
>>pretty pricey software ($100K+). There are 15 engineers who need
>>access to this software. We cannot spend $1.5M to give each
>>engineer a copy, especially when they do not use it all the time.
>>An everyone standing in line to use one machine's console is not
>>practical either. That is where remote logins are necessary.
>
> PMJI, but why is a license server not an option here? No remote login,
> just remote execution...
Ah a VMS user !
The point seems to be that it would be good business for M$ to give us
remote login, if just for the sake of keeping all us large systems people
happy.
DEC (amongst lots of others) have recognised the need for remote login
in large network maintenance, to the point where they impliment remote
login into a Terminal server. (Yes - connect to a terminal server's
console accross the net, test/ configure/ reboot it).
My 2$ worth (Aus $ isn't worth much these days)
...BRU
Path: gmd.de!nntp.gmd.de!Germany.EU.net!EU.net!sunic!trane.uninett.no!eunet.no!
nuug!nac.no!news.kth.se!admin.kth.se!merope.saaf.se!electra.saaf.se!not-for-mail
From: gor...@electra.saaf.se (G|ran Hasse)
Newsgroups: comp.os.ms-windows.programmer.win32,comp.os.ms-windows.nt.misc
Subject: Re: Telnet Daemon for NT
Date: 24 Apr 1994 23:18:46 +0200
Organization: Svensk Amat|rAstronomisk F|rening
Lines: 53
Message-ID: <2pennm$aun@electra.saaf.se>
References: <Cn8nzJ.BHI@tfs.com>
<DON.ALLINGHAM.94Apr12080054@geebee.ftcollinsco.ncr.com>
<tsikesCoMIB6.C16@netcom.com> <1994Apr23.025320.1@cc.curtin.edu.au>
NNTP-Posting-Host: electra.saaf.se
In article <1994Apr23.02532...@cc.curtin.edu.au>,
<zcookb...@cc.curtin.edu.au> wrote:
>In article <tsikesCoMIB6....@netcom.com>, tsi...@netcom.com (Terry Sikes) writes:
>> In article <DON.ALLINGHAM.94Apr12080...@geebee.ftcollinsco.ncr.com>,
>> Don Allingham <Don.Alling...@FtCollinsCO.NCR.COM> wrote:
>>
>> [compress]
>>
>>>One of the big problems we have that requires remote logins is the
>>>cost of software. In my job of circuit design, we tend to use some
>>>pretty pricey software ($100K+). There are 15 engineers who need
>>>access to this software. We cannot spend $1.5M to give each
>>>engineer a copy, especially when they do not use it all the time.
>>>An everyone standing in line to use one machine's console is not
>>>practical either. That is where remote logins are necessary.
>>
>> PMJI, but why is a license server not an option here? No remote login,
>> just remote execution...
>
>Ah a VMS user !
>
>The point seems to be that it would be good business for M$ to give us
>remote login, if just for the sake of keeping all us large systems people
>happy.
>
>DEC (amongst lots of others) have recognised the need for remote login
>in large network maintenance, to the point where they impliment remote
>login into a Terminal server. (Yes - connect to a terminal server's
>console accross the net, test/ configure/ reboot it).
>
>My 2$ worth (Aus $ isn't worth much these days)
>
>...BRU
>
Still i think you are missing ONE god point. I we are going to do
remote execution the software must have all things that needs to be done.
Telnet is for all this things the software producer DID NOT THOUGHT OF.
PLEASE tell me the equivaltent on NT for this..
mailx `grep 100 /etc/passwd | awk -F: '{print $1 }'` < myletter
I bet the NT admins dont even grasp the ideas. Telnet lets the sysadmins
do their jobs in a smart way. The overhead of a graphic userinterface
is a pain. It will kill you in notime!
--
-------------------------------------------------------------
Göran Hasse tel: + 46 8 7730148
Björnmossv 9 fax: + 46 8 7730902
138 37 ÄLTA MIME-mail: g...@cygnus.raditex.se
E-mail: g...@raditex.se
Path: bga.com!news.sprintlink.net!hookup!swrinde!ihnp4.ucsd.edu!munnari.oz.au!
news.uwa.edu.au!nodecg.ncc.telecomwa.oz.au!netbsd08.dn.itg.telecom.com.au!
orca1.vic.design.telecom.com.au!picasso.cssc-syd.tansu.com.au!
wabbit.cc.uow.edu.au!metro!news.cs.su.oz.au!news.adelaide.edu.au!
yoyo.aarnet.edu.au!ntx.City.UniSA.edu.au!levels.unisa.edu.au!ccdps
From: cc...@levels.unisa.edu.au
Newsgroups: comp.os.ms-windows.programmer.win32,comp.os.ms-windows.nt.misc
Subject: Re: Telnet Daemon for NT
Message-ID: <1994May11.135453.22128@levels.unisa.edu.au>
Date: 11 May 94 13:54:53 +0930
References: <Cn8nzJ.BHI@tfs.com>
<DON.ALLINGHAM.94Apr12080054@geebee.ftcollinsco.ncr.com>
<2pennm$aun@electra.saaf.se>
Organization: University of South Australia
Lines: 57
In article <2pennm$...@electra.saaf.se>, gor...@electra.saaf.se (G|ran Hasse) writes:
> In article <1994Apr23.02532...@cc.curtin.edu.au>,
> <zcookb...@cc.curtin.edu.au> wrote:
>>In article <tsikesCoMIB6....@netcom.com>, tsi...@netcom.com (Terry Sikes) writes:
>>> In article <DON.ALLINGHAM.94Apr12080...@geebee.ftcollinsco.ncr.com>,
>>> Don Allingham <Don.Alling...@FtCollinsCO.NCR.COM> wrote:
>>> [compress]
>>>>One of the big problems we have that requires remote logins is the
>>>>cost of software.
>> [compress]
>>The point seems to be that it would be good business for M$ to give us
>>remote login
>> [compress]
>>DEC (amongst lots of others) have recognised the need for remote login
>>in large network maintenance
> [compress]
> Still i think you are missing ONE good point. If we are going to do
> remote execution the software must have all things that needs to be done.
> Telnet is for all this things the software producer DID NOT THOUGHT OF.
> [compress]
> I bet the NT admins dont even grasp the ideas. Telnet lets the sysadmins
> do their jobs in a smart way.
We have a large LAN Manager network to maintain, with some Windows NT. It
is inevitable that we will have lots of Windows NT soon. These issues have
been a worry for two years now.
There are two matters here I think.
1. Remote access to NT boxes
2. Administering NT boxes in a non-interactive way, or at least a
non-GUI way, using scripts.
For remote access, there does not appear to be an immediate out-of-the box
solution. You can buy them from 3rd parties (telnetd's, I mean) or you can
use the rlogin port that has been done from the UCB code. This project needs
some more security added to it though. Part of this issue is that Windows NT
is a single-user OS in terms of concurrent interactive users, but there are
still I am sure lots of people who will be delighted when they can telnet
to their NT box, even if it is into a single login account. The rlogin
solution has another drawback, that nothing displayed by a DOS program is
echoed - I don't know where the output goes in this case.
I am about to cross-post another message about this, but Microsoft don't
seem to be at all convinced that a scripting language in the way most
admins think of it is a necessary thing. I have asked a lot of people who
have influence on the product at various levels and the answers all seem to
some variant of a) use the GUI b) write a service - like Lancaster Uni is
doing for hooking up user registration to a 4GL database thing c) use Visual
basic d) use .BAT files. None of these are terribly satisfactory, even
comparing with not Unix, but Microsoft's own LAN Manager environment.
--
Dan Shearer email: Dan.Shea...@UniSA.edu.au
Information Technology Unit Phone: +61 8 302 3479
University of South Australia Fax : +61 8 302 3385