From Newsgroup: alt.os.multics

In a different forum, someone told the following story:

One of the most clever (in several senses) Multics developers, who was
working in the innermost recesses of the file system, put in a check for a
condition that would be detected during the boot-up sequence -- intended
for himself during debugging/development; because it was an "impossible"
condition, obviously it would never arise in a released, installed system.

So he put in an error message that he could interpret and that amused
him. Neither my memory nor my Latin is good enough to be able to reproduce
the exact wording, but it was a Latin sentence that translated to "Unto the
root this day is born a brother".

I probably don't have to tell you the rest of the story.

Is anyone in the Multics universe familiar with the error message in question, or is it simply folklore?
--
Rich Alderson news@alderson.users.panix.com
Audendum est, et veritas investiganda; quam etiamsi non assequamur,
omnino tamen proprius, quam nunc sumus, ad eam perveniemus.
--Galen --- Synchronet 3.21d-Linux NewsLink 1.2

From Newsgroup: alt.os.multics

Rich Alderson <news@alderson.users.panix.com> writes:

In a different forum, someone told the following story:

One of the most clever (in several senses) Multics developers, who was
working in the innermost recesses of the file system, put in a check for a
condition that would be detected during the boot-up sequence -- intended
for himself during debugging/development; because it was an "impossible"
condition, obviously it would never arise in a released, installed system.

So he put in an error message that he could interpret and that amused
him. Neither my memory nor my Latin is good enough to be able to reproduce
the exact wording, but it was a Latin sentence that translated to "Unto the
root this day is born a brother".

I probably don't have to tell you the rest of the story.

Is anyone in the Multics universe familiar with the error message in question,
or is it simply folklore?

--
Rich Alderson news@alderson.users.panix.com
Audendum est, et veritas investiganda; quam etiamsi non assequamur,
omnino tamen proprius, quam nunc sumus, ad eam perveniemus.
--Galen

I surely won't be the only person to point you to:

https://multicians.org/hodie-natus-est.html
--
Alan Bawden
--- Synchronet 3.21d-Linux NewsLink 1.2

From Newsgroup: alt.os.multics

Rich Alderson <news@alderson.users.panix.com> writes:

[...] it was a Latin sentence that translated to "Unto the
root this day is born a brother".

Is anyone in the Multics universe familiar with the error message in question, or is it simply folklore?

There's an article about it on multicians.org: https://multicians.org/hodie-natus-est.html

The error message doesn't seem to be in the 1982 version of SSTN that
was in the MIT release, although there's a comment in there saying it
was rewritten around 1976: http://web.mit.edu/multics-history/source/Multics/ldd/bos/source/sstn.alm
--
Adam Sampson <ats@offog.org> <http://offog.org/>
--- Synchronet 3.21d-Linux NewsLink 1.2

From Newsgroup: alt.os.multics

On Tuesday, April 7, 2020 at 4:45:33 PM UTC-4, Rich Alderson wrote:

Is anyone in the Multics universe familiar with the error message in question,
or is it simply folklore?

HODIE NATUS EST RADICI FRATER (The console was upper-case only.)

I encountered this message in or before 1975, when I last hacked with the OS.
I believe I was implementing kernel-only processes, which would soon support Peter Bishop's rewrite of page control which cleaned pages in a daemon process instead of in the depths of the page-fault trap. That in turn may have become more important as paging to bulk core eliminated the long latency during which the next page could be freed "for free".

The Latin was over my head; I totally failed to connect the root directory to radix. Luckily I had easy access to Bernie in the next building, and knew his interest in music (and complexity in general) had led him into Latin.

I came away with the impression that the message was in the Salvager rather than BOS, but most importantly, that the immediate cause was that I had not rebuilt the volume management code to match what was then installed on the one-and-only test system. Off peak hours, one (of two) CPUs and a minimum
of memory could be taken off the live time-sharing system, and run with their own IOM, disk, tape, and console. The session would begin with the system already running BOS (booted from tape 0) and the filesystem repaired, so I'd only supply the OS tape (to boot from drive 1).

Another group was just then implementing a major change called VTOC. It would be entirely appropriate for them to make incompatible changes to the filesystem and BOS even if it inconvenienced me. I can no longer say whether I had to merge with uninstalled code but it's likely -- official OS updates were so onerous they only occurred on Saturdays (at which time the system became too
slow to use).

I think merging source, tracking dependencies, and maintaining consistency were entirely manual processes, with no tool beyond diff, and a lot of "pl1 *.pl1". However, the final generation of the OS tape must have let me use a shared library of compiled modules; maybe all I had to do was name the right version. --- Synchronet 3.21d-Linux NewsLink 1.2

From Newsgroup: alt.os.multics

In article <ab338044-0f27-4558-8674-f24f8a15170f@googlegroups.com>,
rfm <rmabee@comcast.net> wrote:

On Tuesday, April 7, 2020 at 4:45:33 PM UTC-4, Rich Alderson wrote:

Is anyone in the Multics universe familiar with the error message in question,
or is it simply folklore?

HODIE NATUS EST RADICI FRATER (The console was upper-case only.)

I encountered this message in or before 1975, when I last hacked with the OS. >I believe I was implementing kernel-only processes, which would soon support >Peter Bishop's rewrite of page control which cleaned pages in a daemon process >instead of in the depths of the page-fault trap. That in turn may have become >more important as paging to bulk core eliminated the long latency during which >the next page could be freed "for free".

I don't think I knew this part of the story (I'm the person Rich was
referrring to who told it in the other group). My recollection is that
the message showed up at a customer site (I'm thinking GM or Ford).
--- Synchronet 3.21d-Linux NewsLink 1.2

From Newsgroup: alt.os.multics

The error message was in the BOS system, an early stage in the bootload process. With Multics release 12.0 (circa 12/86), BOS was eliminated and the error message lost. The earliest known reasonably complete Multics sources are 12.3, plus some fragments from 11.?. Searching those fragments reveals no trace of the error message.
-- Charles
--- Synchronet 3.21d-Linux NewsLink 1.2