On Mar 3 2005, 13:49, Jules Richardson wrote:
> On Thu, 2005-03-03 at 13:09 +0000, Pete Turnbull wrote:
>
> > Probably because Jules' mailer is inserting an unecessary
"Reply-to:"
> > header (pointless, since it's supposed to be used only when you
want
> > replies to go to somewhere other than the "From:")
>
> Possibly fixed now :-) (we'll see when this hits the list...)
Yep, that's fixed it :-)
--
Pete Peter Turnbull
Network Manager
University of York
Hi Bill,
Any chance of your making binary copies of your
working disks available? That would be a boon to the
rest of us OSI owners.
Dave
--- Bill Sudbrink <wh.sudbrink at verizon.net> wrote:
> Well, since my C4P-MF has been rock solid stable
> since I
> fitted the new power supply and cleaned the drive
> head,
> I've been going through a bunch of 25 (and more)
> year old
> diskettes to see just what I have.
>
> Lo-and-behold!
>
> On a diskette simply labeled OS65D 3.2 (not
> originally mine,
> acquired I don't know where), I found really nice
> machine
> code implementations of Space Invaders and
> Asteroids! You
> don't usually see machine code programs on OSI
> diskettes, the
> OS was too crude to have a simple binary loader.
> Diskettes
> usually have BASIC programs, with maybe a couple of
> USR$
> sub-routines in data statements. To load and
> execute the
> programs, you have to EXIT from BASIC into the
> sub-monitor and
> then load the diskette tracks into memory one at a
> time. Once
> you have it loaded, then you GO to the starting
> address. I
> think that these programs might have been originally
> intended
> to be loaded from cassette tape. Fortunately, the
> diskette had
> two BASIC programs, each of which PRINTs the
> instructions for
> loading the machine code programs. I'm really happy
> about this!
> People usually see OSI boxes running rather slow
> interpreted
> BASIC programs. These two programs show just what
> an OSI box
> can do. There is no attribution for the Asteroids
> program, but
> the Space Invaders is copyright 1980 by Michael
> Kincaid.
>
> Can't wait to show these at TCF!
>
> Bill
>
__________________________________________________
Do You Yahoo!?
Tired of spam? Yahoo! Mail has the best spam protection around
http://mail.yahoo.com
Jules Richardson declared on Saturday 26 February 2005 03:03 pm:
> On Sat, 2005-02-26 at 22:02 +0800, Wai-Sun Chia wrote:
> > Born 1966 (age 39).
>
> Most people start at zero ;-)
???
1966 + 39 = 2005. That's the current year, unless during my intoxication
last night, I built a time machine and sent myself back 1 year in
time. :) However, that seems somewhat unlikely.
Pat
--
Purdue University ITAP/RCS --- http://www.itap.purdue.edu/rcs/
The Computer Refuge --- http://computer-refuge.org
>From: msokolov at ivan.harhan.org
>
>Vintage Computer Festival <vcf at siconic.com> wrote:
>
>> This is not your fault. As you may recall, we had this discussion about
>> half a year ago, and it boils down to people not configuring their e-mail
>> clients properly to obey the reply-to directive in the e-mail headers of
>> the list messages.
>
>No, that is not the problem. Since I read my mail with full headers using
>a very "raw" MUA that does *nothing* behind my back, I see what actually
>happens is this: most list messages arrive with Reply-To: set to the list,
>but some arrive with Reply-To: listing both the list and the author.
>
>Since the Reply-To: header is tweaked by the mailing list software, I can't
>see how it can behave so inconsistently. Since I am a hard determinist
>when it comes to computers, the only rational explanation I have is that
>there must be some differences in the headers of messages before they reach
>the list that causes the list software to process them differently. But
>of course in order to pin the problem down, one must see the messages in
>their original form prior to alteration by the list software, which I cannot
>do as a mere subscriber, only Jay or other list staff can do that.
Hi
I just sent myself a message. The only thing I see is that
the reply-to field has my email address in it. The list server
must be simply appending the CCTALK address to what is there.
I don't control that part of my mail tool so I annoy Sellam ;)
Dwight
>
>MS
>
>P.S. One thought: when I was designing my own mailing list management software,
>I implemented a feature by which a list can accept posts from non-subscribers
>with moderator approval. I implemented it so that it set Reply-To: header
>to both the list and the author on those approved outsider posts, on the
>reasoning that the author should see replies but won't see them on the
>list because he is not subscribed. I don't know anything about the software
>Jay uses for this list, but there is a chance that its authors followed
>reasoning similar to mine and the posts arriving with Reply-To: set to the
>list and the author come from non-subscribers approved by moderators.
>
Thanks to a lot of generousity from a lot of folks, the purchase of the new
hardware has been covered as of 1:06pm CST.
Thanks! I'll cut a check to the vendor today.
Jay
Don Hills wrote
>The original diskette design was for a microcode load device for an
>IBM mainframe. I recall there was an IBM Journal of R&D article about
>it, I'll see if I can find it
I happen to have the 25th Anniversary issue of the IBM Journal of Research
and Development, Sept 1981.
There is an article on page 701.
The IBM Diskette and Diskette Drive
by J. T. Engh
The diskette and diskette drive have had a major influence on data
processing. They provide a low-cost, compact, high-performance solution to
the need for a reusable magnetic medium and have largely replaced the
punched card in many applications. Early applications were simple
program-load functions. Today these have expanded to a wide range of medium
exchange, information storage, and data processing applications. This paper
examines the history of the development of these products within IBM. The
discussion includes some of the alternatives considered and some of the
problems encountered during these developments.
You can find it on the IBM Research site
http://www.research.ibm.com/journal/rd/255/ibmrd2505ZE.pdf
My business card was in the book with an email address of
uw-beaver!entropy!dataio!holley
Michael Holley
Next question - anyone think of a good reason why a lot of old tapes I
have containing tar archives seem to hit read errors at around 7.8MB
into the tape?
--
That is the first time the tape reverses direction.
You should always 'retension' the tape before trying to read it
mt -f /dev/st0 retension works under most unix systems.
This runs the tape out, and rewinds to BOT.
If the tapes are old, there is a high probablility the tensioning belt
in the cartridge is going
to break. Keep a few new donor carts around to salvage the belts out of
Be EXTREMELY careful when changing a belt to not damage the tape (it is
very easy to damage
since it is quite thin)
I have a set of five RX50 floppies which contain various service tests. I
have not yet discovered what machine/CPU these work upon. Could anybody
point me in the direction of how to interpret the seven-character part
numbers? All have a copyright date of 1983. They are as follows:
CZUFDB1 (disk part # BL-T540B-M1 "USER TESTS")
CZXD1B1 (disk part # BL-T541B-M1 "FIELD SERVICE TESTS 1")
CZXD2B0 (disk part # BL-T542B-MC "FIELD SERVICE TESTS 2")
CZXD3B0 (disk part # BL-T565B-MC "FIELD SERVICE TESTS 3")
CZXD4B0 (disk part # BL-T583B-MC "FIELD SERVICE TESTS 4")
Are these some version of XXDP? They came with the MVII and MicroPDP-11/73
systems that I got as part of The Great Haul back in July 1998, but there
is no indication who they go to.
I stumbled across something about deciphering those CZX-type numbers a
while back but did not retain the URL (naturally, now that I need to use it).
Thanks much.
-Chris F.
NNNN
Christian R. Fandt, Treasurer
Antique Wireless Association, Inc.
Jamestown, New York USA
email: cfandt at netsync.net
Electronic/Electrical Historian
URL: http://www.antiquewireless.org/
On Fri, 25 Feb 2005 07:12:41 +0100, Nico de Jong <nico at farumdata.dk> wrote:
>
> Yes. The discussion is not about bytes, but about card images.
> A card image (apart from some special ones like 96 bytes and "stubs" of
> about 30 bytes) is 80 bytes. Furthermore, the lowest sectorsize that could
> be formatted on a 3740, was 128 bytes. As there are 73 "user adressable"
> tracks with each 26 sectors, there are only 1898 sectors. And as the
> discussion is about card images, there is only room for 1898 cards, unless a
> technique called spanning is used.
The original diskette design was for a microcode load device for an
IBM mainframe. I recall there was an IBM Journal of R&D article about
it, I'll see if I can find it.
3740 DE (Data Entry) format actually had the option of 80 or 128
characters / "columns" per record. The basic 3741/3742 data entry
station only did 80 characters per record, although the diskette
sectors were always 128 data bytes. You had to order the optional
feature (a plug-in ROM card) to support 128 characters per record.
Track 0 was housekeeping and index. Tracks 1 to 73 ware data. I still
have my 3741 alignment diskette. It has eccentric tracks, and there
was a built-in ROM routine on the 3741/2 station to read a full track
and display the sector addresses in binary. You adjusted the head
alignment to get two areas of good IDs and two areas of bad IDs. The
binary display was unusual, it actually displayed what looked like a
capital "H" with additional horizontal bars for each 1 bit. "00" hex
looked like "H", "FF" hex looked like a solid block,, "18" hex looked
like a square-edged digit "8" etc.
>
> > IBM's 3740 data entry station put its stamp of approval on the
> > floppy. The 3740 format is still the de facto interchange medium
> > within the industry.
> This is worth a whole new topic, as I happen to know a bit about that (I'm
> running a service bureau for media conversion)
>
> > ...
> > IBM's design for the 3740 was very conservative. At the time, IBM
> > believed that floppies would be used for the batch entry of data.
> That is correct; the 3740 family came to be rather big, eventually ending up
> with the 3749(?) data entry station, where one could use DSDD floppies with
> a sector length of 1024. IIRC, it could accommodate 6 segments per track,
> ending up with 8 x 1024 x 73 x 2 = 1.1196.032 bytes
>
> > The full-sized floppy was designed to hold the same amount of
> > information that 3,000 punch cards would hold - the maximum of
> > what a single keypunch operator could do in a day.
> See discussion on "spanning"
>
> Nico
>
>
The recent death of Mr. Raskin has me Google'ing around for information
about his Canon Cat machine.
Even prior to his passing, I have always thought it a sweet looking machine.
Does anyone of this list own a Canon Cat? If so, do you use it for any sort
of regular everyday work? And what's your personal impression of the machine?
Is it even possible to use it in the context of today's work demands (i.e.,
for email)? Or is a proper network interface & appropriate software the
limiting factor here?
