This is G o o g l e's cache of http://www.lp.se/ftp/mailinglists/FREE-VMS.1997-09.
G o o g l e's cache is the snapshot that we took of the page as we crawled the web.
The page may have changed since that time. Click here for the current page without highlighting.
Google is not affiliated with the authors of this page nor responsible for its content. |
| These search terms have been highlighted: | levitte | programming |
|
|
Archive-Date: Wed, 24 Sep 1997 00:21:28 +0200
Sender: <owner-Free-VMS@lp.se>
Message-ID: <342843AD.5D15@net-link.net>
Date: Tue, 23 Sep 1997 18:33:37 -0400
From: "Todd C. Campbell" <toddc@net-link.net>
Reply-To: Free-VMS@lp.se
MIME-Version: 1.0
To: Free-VMS@lp.se
Subject: TEST
References: <7713-Tue20May199722:50:06+0200-levitte@lp.se>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Hello!
I'm checking to see if this still worked...
I've been busy with work, so I have not had much time to
work on anything vms-like until this week...
Starting with setting up some small VAXstation 3100's
...perhaps I will be able to put one on our LAN
at work 10mps to the Chicago Backbone...
If so, I will be able to offer up some workplace
to test new stuff.
*smacks head* I have to look at mach again too...
:)
-todd
================================================================================
Archive-Date: Wed, 24 Sep 1997 18:13:43 +0200
Sender: <owner-Free-VMS@lp.se>
Date: Wed, 24 Sep 1997 12:13:36 -0400
Message-ID: <199709241613.MAA21540@harvey.cyclic.com>
From: Jim Kingdon <kingdonc@cyclic.com>
Reply-To: Free-VMS@lp.se
To: Free-VMS@lp.se
Subject: Re: TEST
> I'm checking to see if this still worked...
Well, I'm still here. I made a new release of VMSBACKUP and mentioned
it here about a month ago. It made it into the mailing list archives,
so I trust people got it.
> If so, I will be able to offer up some workplace to test new stuff.
Unfortunately, there doesn't seem to be any new stuff to test :-(.
Anyone who wishes to take the above statement as a challenge is
encouraged to do so :-).
================================================================================
Archive-Date: Wed, 24 Sep 1997 18:40:50 +0200
Sender: <owner-Free-VMS@lp.se>
To: <free-vms@lp.se>
Subject: Re: TEST
Message-ID: <LEVITTE.97Sep24183520@devil.bofh.se>
From: levitte@lp.se (Richard Levitte - VMS Whacker)
Reply-To: Free-VMS@lp.se
Date: 24 Sep 1997 16:35:20 GMT
References: <199709241613.MAA21540@harvey.cyclic.com>
MIME-Version: 1.0
Content-Type: Text/Plain; Charset=ISO-8859-1
Content-Transfer-Encoding: 8bit
In article <199709241613.MAA21540@harvey.cyclic.com> Jim Kingdon <kingdonc@cyclic.com> writes:
Unfortunately, there doesn't seem to be any new stuff to test :-(.
Yep, and I got pretty busy for quite a while, and going through some
personal trouble, well, that kept me away from the basic stuff I had
to do to live.
One thing that I noticed when things were last discussed was that
except for the occasional blurb about your VMSBACKUP and the start of
the STR$ library clone, all there was was a lot of talk about the
kernel. That is well and fine, but everyone has to understand that
the kernel is the thing that will probably take the absolutelly
longest to build, debug and get ready for use. And there are so many
other things that need being implemented, and can be implemented
TODAY, on your current VMS system. Small petty things like COPY, or
the MTH$ library, or making sure there will be a working TCP/IP stack
(the drivers should be the only parts that should need being rewritten
for new kernels). Or how about an EVE clone?
So, I want to encourage everyone to take some favorite piece of VMS
software (small utilities, larger application, whatever) and start
implementing a clone. Many of those shouldn't even be any difficult
to implement, really...
If what you build works with a current VMS, and doesn't go anywhere
behind the system routines, there's a great chance it will work with
whatever kernel will hopefully show up some day, at least if that work
goes as I wish it to go (source compatibility down to system routine
calls).
If you want to start working on something, please mail me, so I can
set you up on a task list for everyone to look at, so duplicate
efforts will be avoided.
Anyone who wishes to take the above statement as a challenge is
encouraged to do so :-).
I hope my additional blurb will provide an even better challenge, and
a few ideas on how YOU might contribute.
--
R Levitte, Levitte Programming; Spannv. 38, I; S-161 43 Bromma; SWEDEN
Tel: +46-8-26 52 47; Cel: +46-708-20 09 64; No fax right now
PGP key fingerprint = A6 96 C0 34 3A 96 AA 6C B0 D5 9A DF D2 E9 9C 65
Check http://www.lp.se/~levitte for my public key. bastard@bofh.se
"price, performance, quality. Choose any two you like"
================================================================================
Archive-Date: Wed, 24 Sep 1997 21:50:44 +0200
Sender: <owner-Free-VMS@lp.se>
Message-ID: <1.5.4.32.19970924195913.006ac4bc@snake.srv.net>
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Date: Wed, 24 Sep 1997 13:59:13 -0600
To: Free-VMS@lp.se
From: Kevin Handy <kth@srv.net>
Reply-To: Free-VMS@lp.se
Subject: Re: TEST
At 12:13 PM 9/24/97 -0400, you wrote:
>> I'm checking to see if this still worked...
>
>Well, I'm still here. I made a new release of VMSBACKUP and mentioned
>it here about a month ago. It made it into the mailing list archives,
>so I trust people got it.
>
>> If so, I will be able to offer up some workplace to test new stuff.
>
>Unfortunately, there doesn't seem to be any new stuff to test :-(.
>
>Anyone who wishes to take the above statement as a challenge is
>encouraged to do so :-).
If you want to play with the STR$ routines, you should probibly get
the latest version that I have (includes some bug fixes and some
new/untested functions) from my web page. Includes some untested
lib$ functions.
The latest version of BTRAN is also there (updated it just a minute
ago to my current working version). Many fixes/improvements.
http://srv.net/~kth
-------------------------------------------------------------
Kevin Handy kth@srv.net Accounting Software for
Software Solutions. Inc. VAX/VMS Computer Systems
Idaho Falls, Idaho
================================================================================
Archive-Date: Wed, 24 Sep 1997 22:26:10 +0200
Sender: <owner-Free-VMS@lp.se>
Message-ID: <70A95474B7DCD011B58700805F14DD09C52585@RED-70-MSG.dns.microsoft.com>
From: Chris Kaler <ckaler@microsoft.com>
Reply-To: Free-VMS@lp.se
To: "'Free-VMS@lp.se'" <Free-VMS@lp.se>
Subject: RE: TEST
Date: Wed, 24 Sep 1997 13:26:03 -0700
The STR$, LIB$, CLI$, message compiler/rtl are all fundamental libraries
that
can be built on VMS.
As well, getting a core set of SYS$ functions that operate on the pure C
RTL
is also good. This is an approach we took on the prototype and allowed
us to
jump start because Mach came with a BSD implementation.
Getting the basics going on Logical names and event flags is important.
You can initially use a socket-based server to handle the multi-process
synchronization.
Another, somewhat more ambitious, effort would be to build a DCL
emulator.
Christopher
> -----Original Message-----
> From: levitte@lp.se [SMTP:levitte@lp.se]
> Sent: Wednesday, September 24, 1997 9:35 AM
> To: free-vms@lp.se
> Subject: Re: TEST
>
> In article <199709241613.MAA21540@harvey.cyclic.com> Jim Kingdon
> <kingdonc@cyclic.com> writes:
>
> Unfortunately, there doesn't seem to be any new stuff to test :-(.
>
> Yep, and I got pretty busy for quite a while, and going through some
> personal trouble, well, that kept me away from the basic stuff I had
> to do to live.
>
> One thing that I noticed when things were last discussed was that
> except for the occasional blurb about your VMSBACKUP and the start of
> the STR$ library clone, all there was was a lot of talk about the
> kernel. That is well and fine, but everyone has to understand that
> the kernel is the thing that will probably take the absolutelly
> longest to build, debug and get ready for use. And there are so many
> other things that need being implemented, and can be implemented
> TODAY, on your current VMS system. Small petty things like COPY, or
> the MTH$ library, or making sure there will be a working TCP/IP stack
> (the drivers should be the only parts that should need being rewritten
> for new kernels). Or how about an EVE clone?
>
> So, I want to encourage everyone to take some favorite piece of VMS
> software (small utilities, larger application, whatever) and start
> implementing a clone. Many of those shouldn't even be any difficult
> to implement, really...
>
> If what you build works with a current VMS, and doesn't go anywhere
> behind the system routines, there's a great chance it will work with
> whatever kernel will hopefully show up some day, at least if that work
> goes as I wish it to go (source compatibility down to system routine
> calls).
>
> If you want to start working on something, please mail me, so I can
> set you up on a task list for everyone to look at, so duplicate
> efforts will be avoided.
>
> Anyone who wishes to take the above statement as a challenge is
> encouraged to do so :-).
>
> I hope my additional blurb will provide an even better challenge, and
> a few ideas on how YOU might contribute.
>
> --
> R Levitte, Levitte Programming; Spannv. 38, I; S-161 43 Bromma;
> SWEDEN
> Tel: +46-8-26 52 47; Cel: +46-708-20 09 64; No fax right now
> PGP key fingerprint = A6 96 C0 34 3A 96 AA 6C B0 D5 9A DF D2 E9 9C
> 65
> Check http://www.lp.se/~levitte for my public key.
> bastard@bofh.se
>
> "price, performance, quality. Choose any two you like"
================================================================================
Archive-Date: Thu, 25 Sep 1997 00:01:38 +0200
Sender: <owner-Free-VMS@lp.se>
Date: Wed, 24 Sep 1997 18:02:03 -0400
Message-ID: <199709242202.SAA27310@harvey.cyclic.com>
From: Jim Kingdon <kingdonc@cyclic.com>
Reply-To: Free-VMS@lp.se
To: free-vms@lp.se
Subject: web pages
> http://srv.net/~kth
Ahh, two (or more) can play at that game.
For your viewing pleasure:
http://www.cyclic.com/~kingdon/free-vms.html
================================================================================
Archive-Date: Thu, 25 Sep 1997 02:45:30 +0200
Sender: <owner-Free-VMS@lp.se>
To: <free-vms@lp.se>
Subject: Re: web pages
Message-ID: <LEVITTE.97Sep25022937@devil.bofh.se>
From: levitte@lp.se (Richard Levitte - VMS Whacker)
Reply-To: Free-VMS@lp.se
Date: 25 Sep 1997 00:29:37 GMT
References: <199709242202.SAA27310@harvey.cyclic.com>
MIME-Version: 1.0
Content-Type: Text/Plain; Charset=ISO-8859-1
Content-Transfer-Encoding: 8bit
In article <199709242202.SAA27310@harvey.cyclic.com> Jim Kingdon <kingdonc@cyclic.com> writes:
> http://srv.net/~kth
http://www.cyclic.com/~kingdon/free-vms.html
I've added pointers to both at http://www.free-vms.org/activity/tasks.html.
Also, for most entries in that file, I've added a status line, which I
will update as soon as I get more news about the things promised.
That is also the file in which I'll add more entries as required.
--
R Levitte, Levitte Programming; Spannv. 38, I; S-161 43 Bromma; SWEDEN
Tel: +46-8-26 52 47; Cel: +46-708-20 09 64; No fax right now
PGP key fingerprint = A6 96 C0 34 3A 96 AA 6C B0 D5 9A DF D2 E9 9C 65
Check http://www.lp.se/~levitte for my public key. bastard@bofh.se
"price, performance, quality. Choose any two you like"
================================================================================
Archive-Date: Thu, 25 Sep 1997 09:22:26 +0200
Sender: <owner-Free-VMS@lp.se>
Message-ID: <342A112E.2798@netnav.com>
Date: Thu, 25 Sep 1997 02:22:22 -0500
From: "John C. Ellingboe" <jcell@netnav.com>
Reply-To: Free-VMS@lp.se
MIME-Version: 1.0
To: Free-VMS@lp.se
Subject: Re: TEST
References: <199709241613.MAA21540@harvey.cyclic.com> <LEVITTE.97Sep24183520@devil.bofh.se>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Richard Levitte - VMS Whacker wrote:
>
> In article <199709241613.MAA21540@harvey.cyclic.com> Jim Kingdon <kingdonc@cyclic.com> writes:
>
> Unfortunately, there doesn't seem to be any new stuff to test :-(.
>
Snip
>
> TODAY, on your current VMS system. Small petty things like COPY, or
> the MTH$ library, or making sure there will be a working TCP/IP stack
> (the drivers should be the only parts that should need being rewritten
> for new kernels). Or how about an EVE clone?
>
snip
>
> If what you build works with a current VMS, and doesn't go anywhere
> behind the system routines, there's a great chance it will work with
> whatever kernel will hopefully show up some day, at least if that work
> goes as I wish it to go (source compatibility down to system routine
> calls).
>
snip
Hello to the Group.
Richard has mentioned the tcp/ip stack. Has anyone looked at the
CMU-TCP/IP and LIBCMU II package that Clive Nicolson
(clive@bedroom.gen.nz) has put togather. The LIBCMU II is a UCX library
for CMU with socket entry points and UCX$IPC_SHR interface. It does
not, to my understanding, contain the QIO interface yet. I am running
it on my mVAX systems with VMS 5.3 and find that his minor cleanups has
made it much more stable and faster than CMUIP6.6-5K with LIBCMU and
SOCKETSHR which it is based on.
Someone that knows the qio system could probably have it finished up in
a reasonable time frame. I have the sources if any one is interested.
John C. Ellingboe
jcell@netnav.com
================================================================================
Archive-Date: Thu, 25 Sep 1997 10:03:18 +0200
Sender: <owner-Free-VMS@lp.se>
To: <free-vms@lp.se>
Subject: Re: TEST
Message-ID: <LEVITTE.97Sep25095235@devil.bofh.se>
From: levitte@lp.se (Richard Levitte - VMS Whacker)
Reply-To: Free-VMS@lp.se
Date: 25 Sep 1997 07:52:35 GMT
References: <199709241613.MAA21540@harvey.cyclic.com> <LEVITTE.97Sep24183520@devil.bofh.se><342A112E.2798@netnav.com>
MIME-Version: 1.0
Content-Type: Text/Plain; Charset=ISO-8859-1
Content-Transfer-Encoding: 8bit
In article <342A112E.2798@netnav.com> "John C. Ellingboe" <jcell@netnav.com> writes:
Richard has mentioned the tcp/ip stack. Has anyone looked at the
CMU-TCP/IP and LIBCMU II package that Clive Nicolson
(clive@bedroom.gen.nz) has put togather.
Thanks for mentioning that. I knew nothing about LIBCMU II. I've
been thinking about playing with CMU-TCP/IP for some time now, and
yes, I was thinking that one could be used with Free-VMS if possible.
This *will* require reworking whatever device driver that comes with
it, and most possibly parts of the QIO interface, but it's still worth
looking at.
--
R Levitte, Levitte Programming; Spannv. 38, I; S-161 43 Bromma; SWEDEN
Tel: +46-8-26 52 47; Cel: +46-708-20 09 64; No fax right now
PGP key fingerprint = A6 96 C0 34 3A 96 AA 6C B0 D5 9A DF D2 E9 9C 65
Check http://www.lp.se/~levitte for my public key. bastard@bofh.se
"price, performance, quality. Choose any two you like"
================================================================================
Archive-Date: Thu, 25 Sep 1997 18:24:21 +0200
Sender: <owner-Free-VMS@lp.se>
Message-ID: <70A95474B7DCD011B58700805F14DD09FE7F69@RED-70-MSG.dns.microsoft.com>
From: Chris Kaler <ckaler@microsoft.com>
Reply-To: Free-VMS@lp.se
To: "'Free-VMS@lp.se'" <Free-VMS@lp.se>
Subject: RE: TEST
Date: Thu, 25 Sep 1997 09:24:12 -0700
I'm not sure why folks are concerned with an IP stack. I would expect
that to be one of the last things needed. Assuming you build on Mach,
it has a perfectly acceptable FreeBSD IP stack. Why not just "assume"
that IP services will be available. Now building a portable QIO
interface
on top of standard IP socket calls is an interesting and useful project.
Chris
> -----Original Message-----
> From: John C. Ellingboe [SMTP:jcell@netnav.com]
> Sent: Thursday, September 25, 1997 12:22 AM
> To: Free-VMS@lp.se
> Subject: Re: TEST
>
> Richard Levitte - VMS Whacker wrote:
> >
> > In article <199709241613.MAA21540@harvey.cyclic.com> Jim Kingdon
> <kingdonc@cyclic.com> writes:
> >
> > Unfortunately, there doesn't seem to be any new stuff to test
> :-(.
> >
> Snip
> >
> > TODAY, on your current VMS system. Small petty things like COPY, or
> > the MTH$ library, or making sure there will be a working TCP/IP
> stack
> > (the drivers should be the only parts that should need being
> rewritten
> > for new kernels). Or how about an EVE clone?
> >
> snip
> >
> > If what you build works with a current VMS, and doesn't go anywhere
> > behind the system routines, there's a great chance it will work with
> > whatever kernel will hopefully show up some day, at least if that
> work
> > goes as I wish it to go (source compatibility down to system routine
> > calls).
> >
> snip
>
> Hello to the Group.
>
> Richard has mentioned the tcp/ip stack. Has anyone looked at the
> CMU-TCP/IP and LIBCMU II package that Clive Nicolson
> (clive@bedroom.gen.nz) has put togather. The LIBCMU II is a UCX
> library
> for CMU with socket entry points and UCX$IPC_SHR interface. It does
> not, to my understanding, contain the QIO interface yet. I am running
> it on my mVAX systems with VMS 5.3 and find that his minor cleanups
> has
> made it much more stable and faster than CMUIP6.6-5K with LIBCMU and
> SOCKETSHR which it is based on.
>
> Someone that knows the qio system could probably have it finished up
> in
> a reasonable time frame. I have the sources if any one is interested.
>
> John C. Ellingboe
> jcell@netnav.com
================================================================================
Archive-Date: Thu, 25 Sep 1997 19:39:14 +0200
Sender: <owner-Free-VMS@lp.se>
To: <free-vms@lp.se>
Subject: Re: TEST
Message-ID: <LEVITTE.97Sep25193158@devil.bofh.se>
From: levitte@lp.se (Richard Levitte - VMS Whacker)
Reply-To: Free-VMS@lp.se
Date: 25 Sep 1997 17:31:58 GMT
References: <70A95474B7DCD011B58700805F14DD09FE7F69@RED-70-MSG.dns.microsoft.com>
MIME-Version: 1.0
Content-Type: Text/Plain; Charset=ISO-8859-1
Content-Transfer-Encoding: 8bit
In article <70A95474B7DCD011B58700805F14DD09FE7F69@RED-70-MSG.dns.microsoft.com> Chris Kaler <ckaler@microsoft.com> writes:
I'm not sure why folks are concerned with an IP stack. I would expect
Good point.
Now building a portable QIO interface on top of standard IP socket
calls is an interesting and useful project.
:-)
--
R Levitte, Levitte Programming; Spannv. 38, I; S-161 43 Bromma; SWEDEN
Tel: +46-8-26 52 47; Cel: +46-708-20 09 64; No fax right now
PGP key fingerprint = A6 96 C0 34 3A 96 AA 6C B0 D5 9A DF D2 E9 9C 65
Check http://www.lp.se/~levitte for my public key. bastard@bofh.se
"price, performance, quality. Choose any two you like"
================================================================================
Archive-Date: Thu, 25 Sep 1997 21:17:44 +0200
Sender: <owner-Free-VMS@lp.se>
Date: Thu, 25 Sep 1997 15:17:49 -0400
Message-ID: <199709251917.PAA22032@harvey.cyclic.com>
From: Jim Kingdon <kingdonc@cyclic.com>
Reply-To: Free-VMS@lp.se
To: Free-VMS@lp.se
Subject: Re: TEST
> Assuming you build on Mach, it has a perfectly acceptable FreeBSD IP
> stack. . . . Now building a portable QIO interface on top of
> standard IP socket calls is an interesting and useful project.
It is not clear to me that the technical obstacles involved in making
this work are smaller than starting with the CMU IP stack.
More generally, I don't think that we should be shooting down any
project that someone might want to work on. If the response any time
someone start showing some interest in a project is "no, no, no, do it
*THIS WAY*", then I fail to see how anything will get done. If
someone does something the wrong way, the rest of the world can always
just ignore them or smile politely or reimplement it the Right Way or
any number of perfectly adequate responses.
IMHO, of course.
================================================================================
Archive-Date: Thu, 25 Sep 1997 23:40:52 +0200
Sender: <owner-Free-VMS@lp.se>
Message-ID: <199709260746.RAA14278@marvin.technet2000.com.au>
From: "Doug Rickard" <drickard@technet2000.com.au>
Reply-To: Free-VMS@lp.se
To: <Free-VMS@lp.se>
Subject: Re: TEST
Date: Fri, 26 Sep 1997 07:39:51 +1000
MIME-Version: 1.0
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit
> I'm not sure why folks are concerned with an IP stack. I would expect
> that to be one of the last things needed. Assuming you build on Mach,
> it has a perfectly acceptable FreeBSD IP stack. Why not just "assume"
> that IP services will be available. Now building a portable QIO
> interface
> on top of standard IP socket calls is an interesting and useful project.
This had already been done by a Digit some years ago, but was never
released to the public. It allowed a program written for a standard DECnet
interface to run on TCP/IP. It was essentially a DECnet to socket
interface.
I may still have a copy of the sources in my archives. I will search for it
if anyone is really interested.........
Doug.
================================================================================
Archive-Date: Thu, 25 Sep 1997 23:52:02 +0200
Sender: <owner-Free-VMS@lp.se>
Message-ID: <70A95474B7DCD011B58700805F14DD09FE7F6C@RED-70-MSG.dns.microsoft.com>
From: Chris Kaler <ckaler@microsoft.com>
Reply-To: Free-VMS@lp.se
To: "'Free-VMS@lp.se'" <Free-VMS@lp.se>
Subject: RE: TEST
Date: Thu, 25 Sep 1997 14:44:49 -0700
Of course... However, if you want to reach an end-goal,
then you really need a plan of attack that will get you there.
Having been down this road before, my advice is to focus
on the application-level work first. If you implement an IP
stack, it most likely will not plug into the underlying kernel
very nicely. Especially since most successful kernels have
the IP stack inside them. I don't think it is practical to have
people go off and work on kernel pieces until there is a
decision around the kernel and basic architecture. If you
want to do it, great, but my expectation is that it would
not be directly applicable to this effort. IMHO.
As a side note, I don't understand what great technical
obsticals you refer to. We leveraged a great deal of FreeBSD
including the stack, the shell, and the file system.
Christopher
> -----Original Message-----
> From: Jim Kingdon [SMTP:kingdonc@cyclic.com]
> Sent: Thursday, September 25, 1997 12:18 PM
> To: Free-VMS@lp.se
> Subject: Re: TEST
>
> > Assuming you build on Mach, it has a perfectly acceptable FreeBSD IP
> > stack. . . . Now building a portable QIO interface on top of
> > standard IP socket calls is an interesting and useful project.
>
> It is not clear to me that the technical obstacles involved in making
> this work are smaller than starting with the CMU IP stack.
>
> More generally, I don't think that we should be shooting down any
> project that someone might want to work on. If the response any time
> someone start showing some interest in a project is "no, no, no, do it
> *THIS WAY*", then I fail to see how anything will get done. If
> someone does something the wrong way, the rest of the world can always
> just ignore them or smile politely or reimplement it the Right Way or
> any number of perfectly adequate responses.
>
> IMHO, of course.
================================================================================
Archive-Date: Fri, 26 Sep 1997 02:44:49 +0200
Sender: <owner-Free-VMS@lp.se>
Date: Thu, 25 Sep 1997 20:45:19 -0400
Message-ID: <199709260045.UAA24925@harvey.cyclic.com>
From: Jim Kingdon <kingdonc@cyclic.com>
Reply-To: Free-VMS@lp.se
To: Free-VMS@lp.se
Subject: Re: TEST
> Of course... However, if you want to reach an end-goal,
> then you really need a plan of attack that will get you there.
IMHO, that is much more true for a funded project with a tight team
than a distributed free software project. The initial goals of Linux
were nothing like what it ended up being, for example.
Letting/encouraging people to go off in random directions when you are
paying their salaries is one thing (mostly bad, especially beyond a
certain point). Doing it when they are volunteers is quite another,
and instead of spreading the project thin, it tends to increase the
resources available because each volunteer (or their employer, in some
cases) has slightly different reasons for wanting to be involved.
For more thoughts on this, I recommend Eric Raymond's paper _The
Cathedral and the Bazaar_
(http://www.ccil.org/~esr/writings/cathedral.html). For the requisite
flamebait to get you interested I will ask "Why has Linux succeeded
when the GNU Hurd has failed?".
> As a side note, I don't understand what great technical
> obsticals you refer to.
I almost surely lack enough (any) knowledge of CMU TCP/IP internals to
discuss this intelligently. But I am thinking in terms of things like
wanting to be able to read from the network and set event flags and
ASTs, rather than be stuck with the very limited asynchronicity
available with sockets. I may be describing a level of functionality
greater than provided by UCX and Multinet (I don't know), and if so
maybe it isn't very important.
================================================================================
Archive-Date: Fri, 26 Sep 1997 03:28:03 +0200
Sender: <owner-Free-VMS@lp.se>
To: <free-vms@lp.se>
Subject: Re: TEST
Message-ID: <LEVITTE.97Sep26031854@devil.bofh.se>
From: levitte@lp.se (Richard Levitte - VMS Whacker)
Reply-To: Free-VMS@lp.se
Date: 26 Sep 1997 01:18:54 GMT
References: <199709260045.UAA24925@harvey.cyclic.com>
MIME-Version: 1.0
Content-Type: Text/Plain; Charset=ISO-8859-1
Content-Transfer-Encoding: 8bit
In article <199709260045.UAA24925@harvey.cyclic.com> Jim Kingdon <kingdonc@cyclic.com> writes:
For more thoughts on this, I recommend Eric Raymond's paper _The
Cathedral and the Bazaar_
(http://www.ccil.org/~esr/writings/cathedral.html). For the requisite
flamebait to get you interested I will ask
While I agree that the Bazaar model has advantages, I'd say that the
following is much too early to say:
"Why has Linux succeeded when the GNU Hurd has failed?".
Please, let's have further discussion on that topic elsewhere.
> As a side note, I don't understand what great technical
> obsticals you refer to.
I almost surely lack enough (any) knowledge of CMU TCP/IP internals to
discuss this intelligently. But I am thinking in terms of things like
wanting to be able to read from the network and set event flags and
ASTs, rather than be stuck with the very limited asynchronicity
available with sockets. I may be describing a level of functionality
greater than provided by UCX and Multinet (I don't know), and if so
maybe it isn't very important.
As I understand the basic workings of MACH, ASTs shouldn't be too
difficult to implement (through messages?). I dunno how it is with
event flags. And as I understand what Christopher said about this,
TCP/IP is implemented at a pretty low level, with the socket interface
on top. So, I guess it should be possible to implement ASTs and such
by bypassing the socket interface and go for the lowere level
implementation.
--
R Levitte, Levitte Programming; Spannv. 38, I; S-161 43 Bromma; SWEDEN
Tel: +46-8-26 52 47; Cel: +46-708-20 09 64; No fax right now
PGP key fingerprint = A6 96 C0 34 3A 96 AA 6C B0 D5 9A DF D2 E9 9C 65
Check http://www.lp.se/~levitte for my public key. bastard@bofh.se
"price, performance, quality. Choose any two you like"
================================================================================
Archive-Date: Fri, 26 Sep 1997 04:14:44 +0200
Sender: <owner-Free-VMS@lp.se>
Date: Thu, 25 Sep 1997 22:15:15 -0400
Message-ID: <199709260215.WAA11852@harvey.cyclic.com>
From: Jim Kingdon <kingdonc@cyclic.com>
Reply-To: Free-VMS@lp.se
To: Free-VMS@lp.se
Subject: Re: TEST
> "Why has Linux succeeded when the GNU Hurd has failed?".
>
> Please, let's have further discussion on that topic elsewhere.
I said it was flamebait :-). (And it wasn't intended to start a
discussion).
Seriously, I totally agree that discussion of the comparative merits
of different operating systems, or the comparative merits of how
different free software projects run is not appropriate on the
free-vms mailing list (except if there is a _direct_ link to
free-vms).
================================================================================
Archive-Date: Fri, 26 Sep 1997 15:10:44 +0200
Sender: <owner-Free-VMS@lp.se>
From: "christopher f. chiesa" <lvt-cfc@servtech.com>
Reply-To: Free-VMS@lp.se
Message-ID: <199709261310.JAA09553@cyber2.servtech.com>
Subject: Re: TEST
To: Free-VMS@lp.se
Date: Fri, 26 Sep 1997 09:10:26 -0400 (EDT)
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
The idea of putting QIO on _top_ of a TCP/IP stack is ... frightening.
I'd vastly prefer to have QIO first, _then_ implement TCP/IP on top of
THAT...
C.
================================================================================
Archive-Date: Fri, 26 Sep 1997 15:49:46 +0200
Sender: <owner-Free-VMS@lp.se>
To: <free-vms@lp.se>
Subject: Re: TEST
Message-ID: <LEVITTE.97Sep26154932@devil.bofh.se>
From: levitte@lp.se (Richard Levitte - VMS Whacker)
Reply-To: Free-VMS@lp.se
Date: 26 Sep 1997 13:49:32 GMT
References: <199709261310.JAA09553@cyber2.servtech.com>
MIME-Version: 1.0
Content-Type: Text/Plain; Charset=ISO-8859-1
Content-Transfer-Encoding: 8bit
In article <199709261310.JAA09553@cyber2.servtech.com> "christopher f. chiesa" <lvt-cfc@servtech.com> writes:
The idea of putting QIO on _top_ of a TCP/IP stack is ... frightening.
I'd vastly prefer to have QIO first, _then_ implement TCP/IP on top of
THAT...
Have you given that some thought?
--
R Levitte, Levitte Programming; Spannv. 38, I; S-161 43 Bromma; SWEDEN
Tel: +46-8-26 52 47; Cel: +46-708-20 09 64; No fax right now
PGP key fingerprint = A6 96 C0 34 3A 96 AA 6C B0 D5 9A DF D2 E9 9C 65
Check http://www.lp.se/~levitte for my public key. bastard@bofh.se
"price, performance, quality. Choose any two you like"
================================================================================
Archive-Date: Fri, 26 Sep 1997 16:51:16 +0200
Sender: <owner-Free-VMS@lp.se>
Date: Fri, 26 Sep 1997 16:19:49 +0200
Message-ID: <97092616194927@saph1.physik.uni-bonn.de>
From: juengst@saph1.physik.uni-bonn.de (Henry G. Juengst, PIB)
Reply-To: Free-VMS@lp.se
To: Free-VMS@lp.se
Subject: Re: TEST
> The idea of putting QIO on _top_ of a TCP/IP stack is ... frightening.
> I'd vastly prefer to have QIO first, _then_ implement TCP/IP on top of
> THAT...
That is exactly what I though, too.
IMHO the question is, if the new system will be better than the one which
it is based on and depends on, in particular if the basis is something
which one does not like for technical reasons. If the basis is good, then
there would be no real reason to substitute it. Compatiblity is fine and
useful, but on a VMS system the basis should be QIO, not sockets. This is
also true for other parts. Why should someone make a new operating
system, which is in fact just a mask in front of a smudgy system? Will
the result be better?
>
> C.
Henry
================================================================================
Archive-Date: Fri, 26 Sep 1997 18:27:06 +0200
Sender: <owner-Free-VMS@lp.se>
From: "christopher f. chiesa" <lvt-cfc@servtech.com>
Reply-To: Free-VMS@lp.se
Message-ID: <199709261627.MAA19985@cyber2.servtech.com>
Subject: Re: TEST
To: Free-VMS@lp.se
Date: Fri, 26 Sep 1997 12:27:00 -0400 (EDT)
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
[Charset ISO-8859-1 unsupported, filtering to ASCII...]
> In article <199709261310.JAA09553@cyber2.servtech.com> "christopher f. chiesa" <lvt-cfc@servtech.com> writes:
>
> The idea of putting QIO on _top_ of a TCP/IP stack is ... frightening.
> I'd vastly prefer to have QIO first, _then_ implement TCP/IP on top of
> THAT...
>
> Have you given that some thought?
No. It was pretty much a gut reaction after years of enduring the overall
poor design of TCP/IP. I'd never base anything serious "on top of" it.
If I _had_ to have TCP/IP I'd implement it at as high a level of
abstraction as humanly possible, so as to be free to build all levels
BELOW that in a style and architecture superior to TCP/IP's.
Chris Chiesa
lvt-cfc@servtech.com
================================================================================
Archive-Date: Fri, 26 Sep 1997 18:32:49 +0200
Sender: <owner-Free-VMS@lp.se>
From: "christopher f. chiesa" <lvt-cfc@servtech.com>
Reply-To: Free-VMS@lp.se
Message-ID: <199709261632.MAA20261@cyber2.servtech.com>
Subject: Re: TEST
To: Free-VMS@lp.se
Date: Fri, 26 Sep 1997 12:32:44 -0400 (EDT)
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
> > The idea of putting QIO on _top_ of a TCP/IP stack is ... frightening.
> > I'd vastly prefer to have QIO first, _then_ implement TCP/IP on top of
> > THAT...
>
> That is exactly what I though, too.
>
> IMHO the question is, if the new system will be better than the one which
> it is based on and depends on, in particular if the basis is something
> which one does not like for technical reasons. If the basis is good, then
> there would be no real reason to substitute it. Compatiblity is fine and
> useful, but on a VMS system the basis should be QIO, not sockets. This is
> also true for other parts. Why should someone make a new operating
> system, which is in fact just a mask in front of a smudgy system? Will
> the result be better?
Exactly. I would try very hard to implement sockets, TCP/IP, etc. at as
high a level of abstraction as possible, so that as many levels as
possible would be BENEATH that point and could be implemented BETTER than
sockets and TCP/IP.
C.
================================================================================
Archive-Date: Fri, 26 Sep 1997 20:10:57 +0200
Sender: <owner-Free-VMS@lp.se>
Message-ID: <70A95474B7DCD011B58700805F14DD09FE7F70@RED-70-MSG.dns.microsoft.com>
From: Chris Kaler <ckaler@microsoft.com>
Reply-To: Free-VMS@lp.se
To: "'Free-VMS@lp.se'" <Free-VMS@lp.se>
Subject: RE: TEST
Date: Fri, 26 Sep 1997 11:10:10 -0700
There have been several messages regarding the QIO/IP suggestion
I made. It is not my desire to start a flame of philosophical
discussion,
but let me try to explain my thinking behind my last mail.
Clearly there are different approaches and public vs commercial
efforts clearly have different goals and success factors. However,
I would observe that human nature is such that people get distracted
and tend to lose interest. Progress and deliverables engage people
and renew their interest. Consider a presentation of an idea vs a
prototype of the idea. I have been watching this group for quite some
time now and with the exception of a few components, there has
been little progress. This is not meant as a criticism, just an
observation. This is a public effort and we all have jobs that consume
our "non-family" time. However, I would argue that we need to
engage folks. One way to do that is to encourage people to go
off and implement whatever they find interesting. This is a very
good approach for keeping people motivated. However, I beleive
that in this case it has two real downsides: it doesn't help to
engage people who "want to help, but don't know where to start
or how to help". As well, people reach a point where they have
done cool stuff, but they can't pull it all together.
I have suggested before that a plan of attack is to have people
build portable versions of the top layers (libraries, DCL, editors,...).
These can be tested on VMS and used on UNIX (or NT). This
helps to build important pieces, makes progress, and lets people
use their own creativity. However, in parallel, we need a team
to think about what the architecture is going to be. We quickly
get a base system running and then people can fold their suff
onto it. I beleive that to engage people we need to make it easy
for them to add stuff and get immediate feedback from a working
system.
That last thought is was drives the QIO/IP statements that I made.
We could sit down and build a whole new OS from the bottom
up. That is a perfectly valid approach that is propbably very
exciting to many people. However, I believe this will fail because
no one has the time to devote to it and this team is
geographically-challenged.
In our work, a commercial effort, we needed to satisfy managers
that we are making progress. We could have presented reports
with lots of words and numbers to show progress, however, we
decided to let the bits speak for themselves. Being people (well
mostly), the managers became engaged when they saw bits. I
suggest the same thing here. The fact that this is a not a
commercial effort means that people have less time and their
isn't a funding issue, but there is still a motivation issue.
If we decide to adopt a process where we want to have a "FreeVMS"
kit as quickly as possible that people can use, I beleive that we
will see more progress that we have to date. This is especially
key if people can develop software on FreeVMS.
I beleive that this is the best approach and here is how I think
we can do it.
You need to partition the effort in half (I'm sure evryone is
suprised that I said that!). One effort focuses on building the
top layer stuff. The other effort starts with FreeBSD on Mach
and works on the architecture. The top layer stuff can be
build on VMS or on FreeBSD/Mach as it is a fully functional
UNIX system. The top-layer stuff utilizes UNIX initially, but
as the architecture effort makes progress, the top-layer
software can remove UNIX dependencies. This allows for
very quick progress.
As for the specifics of IP vs QIO... These are really two
very different approaches to the same problem. What Mach
will give you (most likely) is a network driver and low-level
protocol stack in the kernel. I suspect that the FreeBSD server
provides the socket mapping layer. My proposal would be
to build (or acquire) portable QIO layers. These can be put on top
of the UNIX file system and sockets initially. This would
allow VMS applications to "run" on FreeVMS. As the
architecture team determines the appropriate low-level services,
the QIO and Socket libraries can change accordingly.
Just my HO... Christopher
> -----Original Message-----
> From: levitte@lp.se [SMTP:levitte@lp.se]
> Sent: Thursday, September 25, 1997 6:19 PM
> To: free-vms@lp.se
> Subject: Re: TEST
>
> In article <199709260045.UAA24925@harvey.cyclic.com> Jim Kingdon
> <kingdonc@cyclic.com> writes:
>
> For more thoughts on this, I recommend Eric Raymond's paper _The
> Cathedral and the Bazaar_
> (http://www.ccil.org/~esr/writings/cathedral.html). For the
> requisite
> flamebait to get you interested I will ask
>
> While I agree that the Bazaar model has advantages, I'd say that the
> following is much too early to say:
>
> "Why has Linux succeeded when the GNU Hurd has failed?".
>
> Please, let's have further discussion on that topic elsewhere.
>
> > As a side note, I don't understand what great technical
> > obsticals you refer to.
>
> I almost surely lack enough (any) knowledge of CMU TCP/IP internals
> to
> discuss this intelligently. But I am thinking in terms of things
> like
> wanting to be able to read from the network and set event flags and
> ASTs, rather than be stuck with the very limited asynchronicity
> available with sockets. I may be describing a level of
> functionality
> greater than provided by UCX and Multinet (I don't know), and if so
> maybe it isn't very important.
>
> As I understand the basic workings of MACH, ASTs shouldn't be too
> difficult to implement (through messages?). I dunno how it is with
> event flags. And as I understand what Christopher said about this,
> TCP/IP is implemented at a pretty low level, with the socket interface
> on top. So, I guess it should be possible to implement ASTs and such
> by bypassing the socket interface and go for the lowere level
> implementation.
>
> --
> R Levitte, Levitte Programming; Spannv. 38, I; S-161 43 Bromma;
> SWEDEN
> Tel: +46-8-26 52 47; Cel: +46-708-20 09 64; No fax right now
> PGP key fingerprint = A6 96 C0 34 3A 96 AA 6C B0 D5 9A DF D2 E9 9C
> 65
> Check http://www.lp.se/~levitte for my public key.
> bastard@bofh.se
>
> "price, performance, quality. Choose any two you like"
================================================================================
Archive-Date: Sat, 27 Sep 1997 00:27:04 +0200
Sender: <owner-Free-VMS@lp.se>
Date: Fri, 26 Sep 1997 18:27:24 -0400
Message-ID: <199709262227.SAA18127@harvey.cyclic.com>
From: Jim Kingdon <kingdonc@cyclic.com>
Reply-To: Free-VMS@lp.se
To: Free-VMS@lp.se
Subject: Re: TEST
In spite of possible apparent disagreements, there is a *lot* of
philosophical agreement between me and Chris (and everyone else I've
heard express an opinion).
For example, I totally agree with the notion that the most important
thing in getting started is coming up with something that people can
download and play with and get instant gratification if they hack on
it. I don't have a strong opinion on what it should be; I don't have
a problem with picking FreeBSD/Mach (I wouldn't have a problem with a
variety of other choices among free OSes either).
My own personal next project probably will be to try out one of the
DCL clones which are already mentioned on the web page, downloading it
and seeing if I can get it to run on linux. If I get around to it I
plan to report my experiences on my free-vms web page and/or this list.
================================================================================
Archive-Date: Sat, 27 Sep 1997 00:49:13 +0200
Sender: <owner-Free-VMS@lp.se>
Message-ID: <70A95474B7DCD011B58700805F14DD09FE7F7B@RED-70-MSG.dns.microsoft.com>
From: Chris Kaler <ckaler@microsoft.com>
Reply-To: Free-VMS@lp.se
To: "'Free-VMS@lp.se'" <Free-VMS@lp.se>
Subject: RE: TEST
Date: Fri, 26 Sep 1997 15:47:50 -0700
I'd like to hear about your experiences with the DCLs. Please e-mail.
I know that I don't generally have time to poll web sites so I like to
have things appear in my in-box.
There are actually two parts to DCL... the language engine itself and
the parser. The parser is quite difficult. We ported most of it to C,
but had to hack the last nine yards. Someone should poll VMS
about the sources. There is a big DECUS this year, someone should
ask why they haven't released the sources to the public domain
at the big VMS Q&A session :-). Getting those would go a long
ways towards getting this project going. There is lots of stuff.
When we do get a DLC going someone should think about compiling
it to a VM. I never got around to that project :-(.
Christopher
> -----Original Message-----
> From: Jim Kingdon [SMTP:kingdonc@cyclic.com]
> Sent: Friday, September 26, 1997 3:27 PM
> To: Free-VMS@lp.se
> Subject: Re: TEST
>
> In spite of possible apparent disagreements, there is a *lot* of
> philosophical agreement between me and Chris (and everyone else I've
> heard express an opinion).
>
> For example, I totally agree with the notion that the most important
> thing in getting started is coming up with something that people can
> download and play with and get instant gratification if they hack on
> it. I don't have a strong opinion on what it should be; I don't have
> a problem with picking FreeBSD/Mach (I wouldn't have a problem with a
> variety of other choices among free OSes either).
>
> My own personal next project probably will be to try out one of the
> DCL clones which are already mentioned on the web page, downloading it
> and seeing if I can get it to run on linux. If I get around to it I
> plan to report my experiences on my free-vms web page and/or this
> list.
================================================================================
Archive-Date: Mon, 29 Sep 1997 07:25:44 +0200
Sender: <owner-Free-VMS@lp.se>
Message-ID: <342F3BD4.11A9@netnav.com>
Date: Mon, 29 Sep 1997 00:25:40 -0500
From: "John C. Ellingboe" <jcell@netnav.com>
Reply-To: Free-VMS@lp.se
MIME-Version: 1.0
To: free-vms@lp.se
Subject: re:test
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
I didn't mean to start a high level controversity by bringing up the
fact that there was alreadt a partially ucx compatiable tcp/ip out
there. The library lacks qio items to be compatiabale with what DEC put
in their ucx library per the maintainer of LIBCMUII.
I thought the idea was to find avaliable resorces to use with FreeVMS to
get the show on the road and an operating system worhing as soon as
practical not to completely reinvent VMS from the ground up to start
with.
John C. Ellingboe
jcell@netnav.com
================================================================================
Archive-Date: Mon, 29 Sep 1997 10:30:43 +0200
Sender: <owner-Free-VMS@lp.se>
Date: Mon, 29 Sep 1997 01:28:48 -0700
From: "Henry W. Miller" <henrym@SACTO.MP.USBR.GOV>
Reply-To: Free-VMS@lp.se
To: Free-VMS@lp.se
CC: henrym@SACTO.MP.USBR.GOV
Message-ID: <009BAFFF.7FD1E50F.129@CVOBKU.CVO.MP.USBR.GOV>
Subject: RE: re:test
> From: MX%"Free-VMS@lp.se" 28-SEP-1997 22:39:41.94
> To: MX%"free-vms@lp.se"
> CC:
> Subj: re:test
>
On Mon, 29 Sep 1997 00:25:40 -0500, "John C. Ellingboe" <jcell@netnav.com> said:
"John C. Ellingboe" <jcell@netnav.com> writes:
> I didn't mean to start a high level controversity by bringing up the
> fact that there was alreadt a partially ucx compatiable tcp/ip out
> there. The library lacks qio items to be compatiabale with what DEC put
> in their ucx library per the maintainer of LIBCMUII.
>
> I thought the idea was to find avaliable resorces to use with FreeVMS to
> get the show on the road and an operating system worhing as soon as
> practical not to completely reinvent VMS from the ground up to start
> with.
>
> John C. Ellingboe
> jcell@netnav.com
One sticky point about using the CMU TCP stack for the Free VMS
project is that the package is written 95%+ in BLISS, with some minor
routines and subsystems in C, Macro, Fortran, Pascal, etc. (E.G. the
device drivers for IP and PTY are written in Macro - they have not been
converted to the VMS 6.1 "Step 2" type drivers that can apparently
contain other high level languages).
Now, I haven't looked at it, but I'm pretty sure that LIBCMU is
written in C - that seems to have been a logical thing to do. But the
underlying stack in still written in BLISS, and despite the fact of
whether you want a QIO or a Sockets interface, (CMU is QIO, BTW) the
stack in not going to run on an arbitrary architecture unless:
1) Somebody rewrites the stack in a more generic language, like C,
or comes up with a BLISS to C translator;
2) Somebody "acquires" a BLISS compiler that'll produce code for
Intel or other architectures.
Note: a number of years ago, DEC marketed a product called PC
ALL-IN-One. Since A1 was written in BLISS for the most part, that would
seem to imply to me that unless someone had rewritten A1 into a more
generic language, there existed a BLISS compiler or cross-compiler for
the Intel architecture that existed at that time, probably just the base
808x processor. Either that, or there was a translator that coverted
BLISS code to Fortran, C or something else. Does anyone have any
information on that? Given the amount of available code for VMS that is
written in BLISS, such a tool could be very useful. Note that I've got
the source code for the PDP-11 BLISS compiler, but it's a cross-compiler
that has to be built first on a PDP-10...
-HWM
================================================================================
Archive-Date: Mon, 29 Sep 1997 14:16:42 +0200
Sender: <owner-Free-VMS@lp.se>
Message-ID: <v03110704b0554c101c93@[199.232.78.21]>
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Date: Mon, 29 Sep 1997 08:16:22 -0400
To: Free-VMS@lp.se
From: Dick Munroe <munroe@acornsw.com>
Reply-To: Free-VMS@lp.se
Subject: RE: re:test
> Note: a number of years ago, DEC marketed a product called PC
>ALL-IN-One. Since A1 was written in BLISS for the most part, that would
>seem to imply to me that unless someone had rewritten A1 into a more
>generic language, there existed a BLISS compiler or cross-compiler for
>the Intel architecture that existed at that time, probably just the base
>808x processor. Either that, or there was a translator that coverted
>BLISS code to Fortran, C or something else. Does anyone have any
>information on that? Given the amount of available code for VMS that is
>written in BLISS, such a tool could be very useful. Note that I've got
>the source code for the PDP-11 BLISS compiler, but it's a cross-compiler
>that has to be built first on a PDP-10...
>
>-HWM
And I had the source code (may still have) a version of Bliss-11 that ran
on RSX-11M in native mode. Digital had it up and running and killed the
project before they could release the compiler; but thats another story for
another time...
Dick
p.s. If you take a look at the CLD via the VERB freeware for the Bliss
compiler on the freeware cds, you'll find a bunch of qualifiers for other
architectures. I've often wondered if these actually WORKED or of they
were just there for planning.
--
Dick Munroe munroe@acornsw.com
Acorn Software (508) 568 1618 x1
267 Cox St. FAX: 562 1133
Hudson, Ma. 01749 http://www.acornsw.com/
Need a web site? a web server? other web service? Contact us...
"They told me to get Windows 3.1 or better so I bought a Macintosh!"
"Then they told me to get Windows 95 or better...so I bought another Mac!"
"Now they tell me to get Windows NT 4.0 Server or better...I've ordered
another Mac!"
================================================================================
Archive-Date: Mon, 29 Sep 1997 18:28:12 +0200
Sender: <owner-Free-VMS@lp.se>
Message-ID: <342FD711.144F@netnav.com>
Date: Mon, 29 Sep 1997 11:28:01 -0500
From: "John C. Ellingboe" <jcell@netnav.com>
Reply-To: Free-VMS@lp.se
MIME-Version: 1.0
To: Free-VMS@lp.se
Subject: Re: test
References: <009BAFFF.7FD1E50F.129@CVOBKU.CVO.MP.USBR.GOV>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Henry W. Miller wrote:
>
> > From: MX%"Free-VMS@lp.se" 28-SEP-1997 22:39:41.94
> > To: MX%"free-vms@lp.se"
> > CC:
> > Subj: re:test
> >
>
> On Mon, 29 Sep 1997 00:25:40 -0500, "John C. Ellingboe" <jcell@netnav.com> said:
> "John C. Ellingboe" <jcell@netnav.com> writes:
>
> > I didn't mean to start a high level controversity by bringing up the
> > fact that there was alreadt a partially ucx compatiable tcp/ip out
> > there. The library lacks qio items to be compatiabale with what DEC put
> > in their ucx library per the maintainer of LIBCMUII.
> >
> > I thought the idea was to find avaliable resorces to use with FreeVMS to
> > get the show on the road and an operating system worhing as soon as
> > practical not to completely reinvent VMS from the ground up to start
> > with.
> >
> > John C. Ellingboe
> > jcell@netnav.com
>
> One sticky point about using the CMU TCP stack for the Free VMS
> project is that the package is written 95%+ in BLISS, with some minor
> routines and subsystems in C, Macro, Fortran, Pascal, etc. (E.G. the
> device drivers for IP and PTY are written in Macro - they have not been
> converted to the VMS 6.1 "Step 2" type drivers that can apparently
> contain other high level languages).
>
All the CMUIP packages I have compile and run with VAXC. I don't even
have BLISS installed on any of my VAXs.
>
> Now, I haven't looked at it, but I'm pretty sure that LIBCMU is
> written in C - that seems to have been a logical thing to do. But the
> underlying stack in still written in BLISS, and despite the fact of
> whether you want a QIO or a Sockets interface, (CMU is QIO, BTW) the
> stack in not going to run on an arbitrary architecture unless:
>
JCE-
================================================================================
Archive-Date: Mon, 29 Sep 1997 18:46:26 +0200
Sender: <owner-Free-VMS@lp.se>
Date: Mon, 29 Sep 1997 09:46:13 -0700
From: "Henry W. Miller" <henrym@SACTO.MP.USBR.GOV>
Reply-To: Free-VMS@lp.se
To: Free-VMS@lp.se
CC: henrym@SACTO.MP.USBR.GOV
Message-ID: <009BB044.FCF5EDE3.129@CVOBKU.CVO.MP.USBR.GOV>
Subject: Re: test
> From: MX%"Free-VMS@lp.se" 29-SEP-1997 09:42:37.61
> To: MX%"Free-VMS@lp.se"
> CC:
> Subj: Re: test
>
On Mon, 29 Sep 1997 11:28:01 -0500, "John C. Ellingboe" <jcell@netnav.com> said:
"John C. Ellingboe" <jcell@netnav.com> writes:
> Henry W. Miller wrote:
> >
> > > From: MX%"Free-VMS@lp.se" 28-SEP-1997 22:39:41.94
> > > To: MX%"free-vms@lp.se"
> > > CC:
> > > Subj: re:test
> > >
> >
> > On Mon, 29 Sep 1997 00:25:40 -0500, "John C. Ellingboe" <jcell@netnav.com> said:
> > "John C. Ellingboe" <jcell@netnav.com> writes:
> >
> > > I didn't mean to start a high level controversity by bringing up the
> > > fact that there was alreadt a partially ucx compatiable tcp/ip out
> > > there. The library lacks qio items to be compatiabale with what DEC put
> > > in their ucx library per the maintainer of LIBCMUII.
> > >
> > > I thought the idea was to find avaliable resorces to use with FreeVMS to
> > > get the show on the road and an operating system worhing as soon as
> > > practical not to completely reinvent VMS from the ground up to start
> > > with.
> > >
> > > John C. Ellingboe
> > > jcell@netnav.com
> >
> > One sticky point about using the CMU TCP stack for the Free VMS
> > project is that the package is written 95%+ in BLISS, with some minor
> > routines and subsystems in C, Macro, Fortran, Pascal, etc. (E.G. the
> > device drivers for IP and PTY are written in Macro - they have not been
> > converted to the VMS 6.1 "Step 2" type drivers that can apparently
> > contain other high level languages).
> >
> All the CMUIP packages I have compile and run with VAXC. I don't even
> have BLISS installed on any of my VAXs.
I'm not certain what packages you are talking about, but the
bulk of the code (IPACP, NAMRES, TELNET, FTP, etc) are written in BLISS.
Many packages that were originally written for other TCP stacks and then
ported to CMU are written in C.
> >
> > Now, I haven't looked at it, but I'm pretty sure that LIBCMU is
> > written in C - that seems to have been a logical thing to do. But the
> > underlying stack in still written in BLISS, and despite the fact of
> > whether you want a QIO or a Sockets interface, (CMU is QIO, BTW) the
> > stack in not going to run on an arbitrary architecture unless:
> >
>
>
> JCE-
-HWM
================================================================================
Archive-Date: Mon, 29 Sep 1997 21:16:20 +0200
Sender: <owner-Free-VMS@lp.se>
Message-ID: <342FFE71.7CA@netnav.com>
Date: Mon, 29 Sep 1997 14:16:01 -0500
From: "John C. Ellingboe" <jcell@netnav.com>
Reply-To: Free-VMS@lp.se
MIME-Version: 1.0
To: Free-VMS@lp.se
Subject: Re: test
References: <009BB044.FCF5EDE3.129@CVOBKU.CVO.MP.USBR.GOV>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
I stand corrected Henry. I went back into the installation backup sets
and there are all sorts of XXX.B files. I learn something new every
day. sorry for the confusion.
JCE-
================================================================================
Archive-Date: Tue, 30 Sep 1997 10:46:06 +0200
Sender: <owner-Free-VMS@lp.se>
Date: Tue, 30 Sep 1997 01:45:48 -0700
From: "Henry W. Miller" <henrym@SACTO.MP.USBR.GOV>
Reply-To: Free-VMS@lp.se
To: Free-VMS@lp.se
CC: henrym@SACTO.MP.USBR.GOV
Message-ID: <009BB0CB.0A55C446.5@CVOBKU.CVO.MP.USBR.GOV>
Subject: Re: test
> From: MX%"Free-VMS@lp.se" 29-SEP-1997 12:29:39.30
> To: MX%"Free-VMS@lp.se"
> CC:
> Subj: Re: test
>
On Mon, 29 Sep 1997 14:16:01 -0500, "John C. Ellingboe" <jcell@netnav.com> said:
"John C. Ellingboe" <jcell@netnav.com> writes:
> I stand corrected Henry. I went back into the installation backup sets
> and there are all sorts of XXX.B files. I learn something new every
> day. sorry for the confusion.
>
> JCE-
John,
That's OK - I'm glad that we were able to get the situation
clarified. FYI, before Bruce Miller left CMU to go to work for TGV, he
was in the process of rewriting the CMU/IP core into C, but that project
obviously got curtailed.
BLISS really is a fun and interesting language to work in. It's
too bad that it is not available on other platforms. When I was at SRI,
and we were in the process of converting the ARPANET from NCP (The
original protocol, not at all related to DEC's NCP, but in some ways
similar to ANF-10) to TCP, I wrote an SMTP client/server in a single
morning using BLISS-10, so that Paul Mockapetris and Marc Crispin would
have someone to talk to. It is, obviously, very well suited for systems
work.
-HWM
================================================================================
Archive-Date: Tue, 30 Sep 1997 11:15:34 +0200
Sender: <owner-Free-VMS@lp.se>
Date: Tue, 30 Sep 1997 02:15:28 -0700
From: "Henry W. Miller" <henrym@SACTO.MP.USBR.GOV>
Reply-To: Free-VMS@lp.se
To: Free-VMS@lp.se
CC: henrym@SACTO.MP.USBR.GOV
Message-ID: <009BB0CF.2F08AA34.37@CVOBKU.CVO.MP.USBR.GOV>
Subject: RE: re:test
> From: MX%"Free-VMS@lp.se" 29-SEP-1997 05:28:28.46
> To: MX%"Free-VMS@lp.se"
> CC:
> Subj: RE: re:test
>
On Mon, 29 Sep 1997 08:16:22 -0400, Dick Munroe <munroe@acornsw.com> said:
Dick Munroe <munroe@acornsw.com> writes:
Dick,
> > Note: a number of years ago, DEC marketed a product called PC
> >ALL-IN-One. Since A1 was written in BLISS for the most part, that would
> >seem to imply to me that unless someone had rewritten A1 into a more
> >generic language, there existed a BLISS compiler or cross-compiler for
> >the Intel architecture that existed at that time, probably just the base
> >808x processor. Either that, or there was a translator that coverted
> >BLISS code to Fortran, C or something else. Does anyone have any
> >information on that? Given the amount of available code for VMS that is
> >written in BLISS, such a tool could be very useful. Note that I've got
> >the source code for the PDP-11 BLISS compiler, but it's a cross-compiler
> >that has to be built first on a PDP-10...
> >
> >-HWM
>
> And I had the source code (may still have) a version of Bliss-11 that ran
> on RSX-11M in native mode. Digital had it up and running and killed the
> project before they could release the compiler; but thats another story for
> another time...
>
> Dick
>
> p.s. If you take a look at the CLD via the VERB freeware for the Bliss
> compiler on the freeware cds, you'll find a bunch of qualifiers for other
> architectures. I've often wondered if these actually WORKED or of they
> were just there for planning.
>
VERY interesting... But, for example, the command:
$ BLISS/INTEL
wants to invoke the image BLISS32IW.EXE, which was not
installed, at least, with the installation of BLISS that I did. Are
these cross compilers available anywhere?
-HWM
> --
> Dick Munroe munroe@acornsw.com
> Acorn Software (508) 568 1618 x1
> 267 Cox St. FAX: 562 1133
> Hudson, Ma. 01749 http://www.acornsw.com/
>
> Need a web site? a web server? other web service? Contact us...
> "They told me to get Windows 3.1 or better so I bought a Macintosh!"
> "Then they told me to get Windows 95 or better...so I bought another Mac!"
> "Now they tell me to get Windows NT 4.0 Server or better...I've ordered
> another Mac!"
>
>
================================================================================
Archive-Date: Tue, 30 Sep 1997 13:48:39 +0200
Sender: <owner-Free-VMS@lp.se>
Date: Tue, 30 Sep 1997 12:45:34 +0200
Message-ID: <97093012453438@krysia.uni.lodz.pl>
From: pwierzb@krysia.uni.lodz.pl
Reply-To: Free-VMS@lp.se
To: FREE-VMS@lp.se
signoff