I made a brief and vain attempt to search for this subject on
the web. Maybe when search engines are better at parsing a
request than they are at crawling...
Lots of software I've seen over the last 25 years includes
some text file of instructions with a name chosen to grab
a person's attention and get them to read that document first.
The event-horizon of these documents is not what I'm curious
about, but rather, the particular choice of name... the name(s)
I mention in the subject line are all self-referential, which
always elicits a giggle from me.
I'm betting some document accompanying one of the PDP-8
operating systems or software packages will probably be
But does anyone know? Has this subject been discussed before?
-Douglas Hurst Quebbeman (DougQ at ixnayamspayIgLou.com) [Call me "Doug"]
Surgically excise the pig-latin from my e-mail address in order to reply
"The large print giveth, and the small print taketh away." -Tom Waits
I'm not sure exactly when they started but IBM sent a document called an
"Application Directory" that contained last minute info and a description of
what was on the media - very close to a readme file. I remember them from
the mid to late sixties and they were still used at least until 1992
(Probably until the present). The name isn't very catchy, though............
The oldest I found in the PDP-10 archives is the UCI LISP
"READ.ME" from the 4th DECUS library tape, with a timestamp of
Unfortunately, many of the earlier mini OS's do not support a
[unique] timestamp in the tape and/or disk directory structure,
so it's hard to tell when a document was really last edited.
True, but if it's part of a complete software release, there
may be ancilarry documents that contain dates int he text that
would help place them temporally...
Our standard didn't call them READ.MEs. We found that customers
didn't read them first. We called them *.BWRs, short for beware.
>Unfortunately, many of the earlier mini OS's do not support a
>[unique] timestamp in the tape and/or disk directory structure,
>so it's hard to tell when a document was really last edited.
Even if files did have a time/stamp, it still wouldn't tell
you last edit. That's why we always put an edit history in
the sources. Packaging procedures' date stamps would override
any editing datestamp.
You can tell publication releases by looking at the update history
(but that's just for Software Notebooks). If you're looking at
a stand-alone manual, you can get clues based on the printing
histories and copyright dates.
Everything else was stored in DEC's internal doc retrieval system.
I suspect all of that stuff is long gone.
Subtract a hundred and four for e-mail.
Perhaps the idea originated with Lewis Carroll?
Good chance that the author of the first so-named file
was a Carroll fan...
Probably not. One of the document specs (yes, we had
specs on how to format documents of all flavors...manuals had
one flavor, functional specs had another flavor, memoes
had yet another...) did have us changing the date in the file
to the date it was getting edited; but these were only
functional specifications. I believe examples can be found
in [emoticon gets up and looks at all of its -10 notebooks]
shit...I can't find it. There was a section of a notebook,
perhaps #14?, that had TOPS-10's functional specs. The first
page had three lines, the name of file, the date, and I think
the version number (which was really a document revision number).
Never, AFAICR, did I ever type a file whose name was READ.ME
or any variation. It was not unique enough. Now, they may have
easily renamed a file to that; I never saw the final packaging
Probably unrelated, but worth it nonetheless:
I was setting up a networked laserprinter, but for some reason the standard
tools didn't always flush the last page from the printer. So I modified the
print queue so each job had 60000 "nonprintable" characters appended to the
end, enough to always ensure a flush.
But what to choose for the "nonprintable" characters? ASCII NUL's didn't work,
they were already being stripped out by some other layer of the printing
tools. Ahah - ASCII BEL's didn't get stripped, and they had no apparent
side-effects on the laserprinter. So I made a file with 60000 BEL's and
named it "DONT_TYPE_ME.DAT". Do some more tests, everything looks good,
so I changed the MOTD so that everyone would know about the availability
of the new printer.
Five minutes later, from down the hall, I hear a stream of continuous beeps...
A cow orker had taken a peek at my quick hack :-).
Moral: nobody may read your READ.ME. But everyone will type your DONT_TYPE_ME.
>Moral: nobody may read your READ.ME. But everyone will type your
That's like the old rule: If you want people to touch something,
put a sign on it saying "WET PAINT".
My favourite READ.ME file, at least from the standpoint of
catching snoops, contains a single line saying "File not found".
cgi...@sky.bus.com (Charlie Gibbs)
Remove the first period after the "at" sign to reply.
I don't read top-posted messages. If you want me to see your reply,
appropriately trim the quoted text and put your reply below it.
That can be caused by a wayward finger typing a command to
PIP and a combination of other happenstances. I can't remember
the details of how I got myself in that jam, but it took me
the longest damned time to figure out the file was really there
and all it was seeing was its contents. Bless TW for his
SET WATCH FILE implementation.
Line up ten people along a rope and shout 'pull'. Five of them
will push and two will say, 'What?'
--- Terry Pratchett (approximately)
+- David Given --McQ-+ "For is it not written, wheresoever two or three are
| d...@cowlark.com | gathered together, yea they will perform the Parrot
| (d...@tao-group.com) | Sketch?" --- _Not The 9 o'Clock News_
+- www.cowlark.com --+
Traditionally, APL workspaces have a variable named DESCRIBE.