1 Jun
2023
1 Jun
'23
2:21 a.m.
A VMS listing on Bitsavers? That will be worth a look, for sure. I
guess it would also include the NICE of the era?
Yes, I know Bliss. Or knew it... Bliss was a huge deal in Marlboro in
the late 70's when I was around or at least corporate was trying to make
it so. Any new hire was signed up for a Bliss Common class where the
all the virtues of cross platform compilation were extolled. Bliss
wasn't adopted as a replacement for any existing major 36 bit software
system that I recall, with a few exceptions. So, I don't recall that
any Tops-20 monitor modules were replaced. Nothing in the EXEC was, but
the EXEC's PCL interpreter was written in Bliss.
As I recall, Bliss was used in the implementation of certain languages.
Not for Algol, which has one of the more interesting macro sources I've
seen. FORTRAN is.
Although somewhat cross platform (I.E. Tops-10 and Tops-20), Galaxy
remained in assembler with the exception of the NICE process (NMLT20).
This was originally written in assembler, as can be seen from the PHASE
II Tops-20 DECnet sources but was eventually implemented in Bliss and
not released. This caused a fair amount of ill will as early Bliss
versions had stability issues and crashed regularly. I typed up a
number of SPR's about that.
It would be an interesting question to see how much code was actually
shared for NICE; one assumes it could have been a fair amount. The
Tops-20 RMS-20 product appears to have a DAP implementation in BLISS, so
that will be worth looking at.
I think Bliss probably had greater platform penetration for VMS. The
Kermit-10 is actually written in Bliss and will compile for both Tops-10
and VMS. In fact, because it uses Galaxy, it has compilation hooks for
Tops-20, a number of which are implemented. I think it might be
interesting to play around with that at some point as it appears to be
an orphan now.
It's a shame about the lost specs, particularly if they were adhered to,
which appears to largely be the case with DECnet. For certain Internet
protocols, you just have to see what is going over the wire and reverse
engineer from that. Not the end of the world, but not particularly
pleasant, either.
------------------------------------------------------------------------
On 5/31/23 7:31 PM, Paul Koning wrote:
I have never seen one, which is odd because obviously both 6.0 and 7.0 were implemented
and therefore must have existed. There are various other DECnet Phase IV era specs that
appear to be lost, for example the IV-plus routing spec. I implemented bits of that in
PyDECnet from memory (hopefully correctly...) but there are other things, like congestion
control and equal cost path splitting, that I sure would like to see specified.
There is a VMS 4.x listing set on Bitsavers (scanned microfiche) which I think includes
DAP 7.0 code. I spent a little time reading through pieces of that when I was working on
the DAP code in PyDECnet. So you might try that; chances are you know Bliss, which I do
not. And/or you can look at the PyDECnet code, though of course then you'd be dealing
with second generation reverse engineering.
> ------------------------------------------------------------------------
> On May 31, 2023, at 7:21 PM, Thomas DeBellis<tommytimesharing(a)gmail.com>
wrote:
>
> I finally put something together that is analogous to the RSX NFT /ID switch.
Basically, it grabs a copy of the configuration message before negotiation and displays it
later. So for VENTI2::, one sees:
> ...
>
> Gee, MIM:: does a lot of stuff, so I regrettably remain with a pretty bad case of
data envy... And I really have to get off my butt and implement rename.
>
> Is there an ASCII version of DAP 7.1 anywhere on HECnet that I could download? I was
curious about file preallocation whose bit is defined in the 5.6 document, but not really
explained, as far as I saw.