Jochen Kunz wrote:
...
>I see two problems:
>1. Bus transciever chips.
Yes, this is the big one. It turns out to be solvable, but not using
IC's. National DS3862 would be good, but it just went out of production...
I'm looking into making a "trapzoidal driver" (i.e. controlled edges)
using a FET and RC on the gate. Someone else suggested it and it
sounded like a good idea. Certainly easy to model/simulate first.
-brad
Hi,
I asked this on the classic computer list and I thought I'd ask here
also...
Does anyone have any thoughts on how hard it would be to make a unibus
board which is an IDE controller?
I have 4-6 layer boards fabbed regularly and use modern CPLD's & VHDL on
a regular basis, so the building part looks easy.
I've never looked at unibus controlleqr schematic, but plan to. I'm
assuming much of the old ttl can be sucked into something like a Xilinx
coolrunner CPLD...
I also assume it's reasonably straightforward TTL, and at (by today's
standards) slow speed... true?
Any hints, or gotcha's as far as fabrication or interface? Has anyone
done this (in the modern day, that is :-)
My plan would be to build a 4 layer board of suitable thickness with
gold fingers, using an existing board for reference (any physical size
specs I could read?)
I'm well aware of the foolishness of this on one level, but there's a
side of me that really enjoys this sort of thing... perhaps medication
would help :-)
-brad
medialab.dyndns.org blocked - medialab.freaknet.org is the solution!
dyndns.org blocked our domain "medialab.dyndns.org" for a not
well specified "account violation" problem. we're trying to solve
that, but the real problems is the large amount of google indexes
pointing to medialab.dyndns.org
this free domain was born when we was not able to register a "real"
domain. now we have freaknet.org and medialab.freaknet.org, but there's
nothing to do for all the old links resting into the net :)
this address is used from many users around the world, to telnet,
ssh, rlogin and surf our free computer network!
so, here i announce that medialab.dyndns.org now is not working
anymore, and people can use medialab.freaknet.org instead.
Hope this message will be soon indexed by google :)))
sorry for this - more details and all the story background can be
read at our main site, http://www.freaknet.org in the news section.
tnx all and god bless all PDP/11 in the world.
--
[asbesto : freaknet medialab : radio#cybernet : GPG key on keyservers]
[ MAIL ATTACH, SPAM, HTML, WORD, and msgs larger than 95K > /dev/null ]
[http://www.freaknet.org/asbesto IW9HGS http://kyuzz.org/radiocybernet]
This one was too good to not pass around.
http://moneycentral.msn.com/investor/srs/srsmain.asp?Symbol=SCOX
The sites says (and I quote)
<< The SCO Group, Inc., a small-cap growth company in the technology sector, is
expected to significantly underperform the market over the next six months with
very high risk. >>
__________________________________
Do you Yahoo!?
Yahoo! Hotjobs: Enter the "Signing Bonus" Sweepstakes
http://hotjobs.sweepstakes.yahoo.com/signingbonus
norman(a)nose.cs.utoronto.ca (Norman Wilson) wrote:
> You mean you've restored the original version of cat that had only one option,
> [...]
Notice my use of the words "nearly" and "almost" in the part you responded to.
Seriously though, you gotta agree that until 4.3BSD inclusive, Berkeley was
basically adding to and extending V7. Sure they added a *lot* and extended
many of the existing facilities, but with very few exceptions, it was all
additive, virtually no V7 facility (except the mpx you mentioned) was removed.
Yes, they added fsck, but icheck is still there! (No one uses it of course,
but knowing that nobody removed it gives a warm fuzzy feeling.) The same goes
for almost everything else.
Here is the acid test: time-teleport a V7 user from 1979 to a VAX running
4.3BSD, set PATH=/bin:/usr/bin (no /usr/ucb), do stty old (old tty driver) and
stty ek (erase # kill @) and see if he feels at home or not.
Of course I never use my systems in this way, I make extensive use of Berkeley
UNIX facilities, but I like it much better to use a system that is additive
rather than substitutive with respect to Original UNIX.
MS
Michael Sokolov:
It feels so great knowing that my current modern OS (last release
2003-12-07 counts as current and modern to me) still has nearly all original V7
UNIX code almost completely untouched. It's what gives me the right to call it
UNIX.
=======
You mean you've restored the original version of cat that had only one option,
and the version of ls that had fewer than a dozen and didn't care how wide
the screen was; that filenames are only 14 characters long; that fsck has
been abolished in favour of icheck and ncheck and dcheck; and that file system
blocks have returned to their original V7 size of 512 bytes?
My hat's off to you if so.
On the other hand, I have to question either your stability or that of your
system if you have reinstated the original V7 code implementing mpx(2).
Norman Wilson
Toronto ON
(who actually used the old multiplexor once, but had to fix it first!)
Kenneth Stailey <kstailey(a)yahoo.com> wrote:
> Warren, I see that there is a 4.4BSD-Alpha subdir in the TUHS archive. Do you
> want a final CSRG 4.4BSD tape to add there too?
I would definitely want an image of the 4.4BSD tape too! But I do mean *tape*,
not what's on Kirk's CD-ROM. I'm talking about a record-for-record image of
the Official Release Master tape.
> Thanks, I have to go through your archive for it now.
Actually I just took a closer look and it has been in the distributed
/usr/lib/learn since 4.3BSD, it was just never added to /usr/src/usr.lib/learn
for some reason (not even in Quasijarus, I probably didn't notice it). So you
don't have to go through the pain of downloading a dist from Harhan to pull it
out of there, you can just take it from your CD-ROM set.
> It's so difficult being you.
:-)
> I was able to save myself the time by
> using the 4.4BSD version of learn(1) since it has already been
> modified for CRT terminals. You will have to re-invent the wheel
> because of your politics.
Well I'll take a look at what they did to learn in 4.4BSD and see if any of it
is acceptable for Quasijarus. Hopefully I won't have to reinvent the wheel.
I believe in adding new features without breaking or disturbing historical
stuff. It feels so great knowing that my current modern OS (last release
2003-12-07 counts as current and modern to me) still has nearly all original V7
UNIX code almost completely untouched. It's what gives me the right to call it
UNIX.
As far as learn goes I think I would need to add new lessons for UNIX on a CRT
or some options or somesuch, but I do NOT want to remove the facility for
teaching UNIX on a hardcopy tty. That's such a gem, it should be kept!
MS
Kenneth Stailey <kstailey(a)yahoo.com> wrote:
> When I purchased my copy of _The
> CSRG Archives_ CDROM set I was told in E-mail by McKusick that I did not need
> to sign any license agreements. I am assuming that this is due to Caldera
> proclaiming that V32 sources and binaries could be redistributed by the public
> legally.
Yes, this is correct, this is the same reason why Warren was able to remove the
password system from his UNIX Archive and make it completely open.
> In the CDROM set is a fully encumbered 4.4BSD source tree which
> includes the learn(1) source code.
Yup, I have it too (the whole CD-ROM set). learn(1) is far older than 4.4BSD
though, and goes way back. 4.3BSD-Quasijarus has it too.
> I also have yet to
> get the vi lesson data which the source code that I do have says came on a
> separate user-contributed tape.
I just looked and 4.3BSD-Quasijarus has the vi lesson data as part of the
standard system.
> I got here because I have newbies in my life now and I need UNIX online
> courseware.
Hear hear. I sometimes get into this situation too, usually when dating and
getting faced with the need to teach a prospective female how to use a real
operating system, since the one woman who finally makes it would absolutely
have to use 4.3BSD-Quasijarus on my VAXen.
I looked into learn, but one thing it disappointed me with is that it's
woefully outdated. It starts by setting the tty erase and kill chars to '#'
and '@' respectively and teaching you how to edit the command line on a
hardcopy tty. Well, OK, some would see this as good educational value, but the
problem is, if you don't actually *have* a hardcopy tty, and most of us don't,
it doesn't work too well. It prints out lessons longer than 24 lines and they
scroll off the top of the VT terminal. It was definitely written with the
assumption that one has a hardcopy tty with a long roll of continuous paper,
and it expects the student to grab the paper coming out of the teletype and
look at what's been printed, but it just doesn't work on a VT terminal. Not to
mention that in the end the lessons give the student little practical learning
that would actually be useful when using UNIX on a CRT terminal. (For example,
it would be very practical to explain to the student the difference between ^H
and ^? and teach him/her how to deal with it.)
> I'm wondering about distributing the results of my porting effort once it
> matures enough to be worth doing so.
Well, as a I said 4.3BSD-Quasijarus contains learn and all other "encumbered
code" and it is freely available via anonymous FTP from ifctfvax.Harhan.ORG,
so... BTW for those who missed it I released 4.3-QJ0b on 2003-12-07.
MS
> I would be interested to know if there is any chance of getting a hobbyist
> A/UX license out of Apple.
A/UX started out as a port of Unisoft SysV. Prior to version 2.0, there was
no Finder interface at all. A real history of the product should be done at
some point, as opposed to the half-baked opinions of someone who has only
seen a very late version of the product.
Since there were per-copy licensing fees to Unisoft, at least for the early
versions, it seems unlikely that a hobbyist license for A/UX would be possible.
