Re: LYNX-DEV Problems installing Lynx on VAX...

[Hiram Lester, Jr.]

On Thu, 12 Dec 1996, Foteos Macrides wrote:

>       Don't equate "VAX" with "VMS".  VMS is an operating system. A
> VAX is hardware.  DEC developed the alpha chip, a RISC processor, which
> is used in their AXP systems (for both VMS and OSF) as well as in many
> non-DEC Unix systems.  VMS can be used on both VAXen (which have CISC
> processors, and are slow compared to RISCs) and AXPen.  We got rid of
> all our VAXen and MicroVAXen (and MINCs, PDPs, etc. 8-) and replaced
> them with an AXP 2100.

Indeed.  It used to be good enough to refer to DEC machines as VAX's
because the VAX was their flagship, but no more.  I think the VMS 5.5
welcome string was "VAX at your service on node blah", and now it's "VMS
at your service...".  I remember several changes just recently.  There's
also the more ANSI compliant DEC C to replace VAX C, so they are phasing
out the VAX name for general usage to refer to their machines.  I don't
know if any of this relates to the change from the name "Ultrix" to "DEC
Unix" for their unix clone (Ultrix was VERY BSDish even down to the
version numbers, not sure about DEC Unix).

> as well, except for the mods zip here, which Hiram checks out carefully on
> Unix and at intervals adds to the hopefully "production quality" Composite
> updates for both Unix and VMS. 

Yes, and since I check it so carefully, I haven't had time for it during
finals, but they are drawing to a close, so I'll probably get to it really
soon... :)  I *MIGHT* have time to do something tonight, although this
weekend seems more reasonable.

> I should also note that it's been a *very*
> long time since any VMSer besides me has contributed to Lynx.  In my
> experience, VMSers are even worse at expecting "an expert" to do it for
> them instead of rolling up their sleeves and giving it a shot themselves.

I'd love to do more with VMS.  My first intro to Lynx was one of the
earliest VMS versions when WWW sites were still few and far between and
Mosaic was king (not even sure if Netscape existed at the time).  Any idea
what version that might have been? (around Spring '94?)

My major problem is that I don't have access to a VMS system.  I still
have an account on the gopher/web/news server at my old school because I'm
listed as the contact for an organization's gopher menus, but I don't
really want to use the account much since it could go away at any moment,
and they might get upset with me if they find me using it. (I worked with
these people when I was there.)  When I tried to compile lynx once before
there on one of the other machines, I ran out of virtual memory and was
unable to get the executable built!  I have built several VMS binaries,
but not lynx...

I've done a bit of VMS programming, but not much of it truly VMS specific
other than using the VMS error message routines to provide true VMS-style
errors in one program.  I know NOTHING about VMS's convoluted ideas about
sockets. ;)

> Also, quite some time ago,
> Andy Harper added SOCKETSHR support for Lynx, which I've tried to
> keep up.  The SOCKETSHR library is available as a binary, and if
> built for that, there should be no problem with variations in TCPIP
> packages on different VMS systems.  I.e., it in theory created the
> situation in which one VMS binary could be used by all VMSers, though
> no one has followed through on that.

I'm not sure about binary compatibility, but NETLIB by MadGoat software at
Western Kentucky should be enough to smooth over the differences between
various TCP/IP implementations.  SOCKETSHR just adds another layer to
NETLIB and provides a more unix-like socket interface for VMS across the
various platforms.  SOCKETSHR should make porting unix programs to VMS a
much simpler task...  I've never actually used either one though (other
than incidentally using it when compiling Newsrdr also by MadGoat.

   +------------------------------------+-------------------------------+
   | Hiram W. Lester, Jr.               | E-Mail: address@hidden    |
   | Computer Science                   | Home page:                    |
   | Middle Tennessee State University  |   http://pobox.com/~hwlester/ |
   +------------------------------------+-------------------------------+

;
; To UNSUBSCRIBE:  Send a mail message to address@hidden
;                  with "unsubscribe lynx-dev" (without the
;                  quotation marks) on a line by itself.
;