UZ32112%SG2.BITNET@WISCVM.WISC.EDU (08/10/86)
Date: 8 August 1986, 16:49:37 ULG
From: Andre PIRARD +32 (41) 520180(449) UZ32112 at SG2
SEGI - Universite de Liege
15, av. des Tilleuls
B4000 LIEGE
Belgique
To: INFO-MIC at BRL
Hi!
We, BITNET users, have no interactive access to Arpanet and related nets.
Our only possibility is mailing. Consequently, no FTP.
Imagine our frustration when reading all that fuss about which
free software is the best when we can't even get a single one.
From several sources, I was recommended to find a key pal to which I
would ask mailing things. But everyone imagines this is no long range
solution as it would bother otherwise very busy people.
It would however be conceivable to easily retrieve data by mail.
The server solution is inadequate because a server has to be installed
in each site owning a database.
On the other hand, there are other ways to achieve the same results by
installing server-like interfaces on one or several sites offering the
service. They could, in the simplest case, take the incoming mail and
feed it as input to interactive sessions. The session output would just
be mailed back. A macro system could turn the systen to be responsive
to simple case commands.
I forcedly have no experience of the details of those particular interactive
systems and can't evaluate the feasability of my sayings, but I imagine
it would not be less usable that some mail based databases I could try.
Now my questions are:
1) does anyone know of such already existing system?
2) if not, are there other forums or any person that would be interested
by the idea? (Arpanet might also benefit from the system by
having the possibility to delay heavy transfers during the weak hours).
[sorry no joke] A. PirardUZ32112%SG2.BITNET@WISCVM.WISC.EDU (09/02/86)
Date: 2 September 1986, 12:58:17 ULG
From: Andre PIRARD +32 (41) 520180(449) UZ32112 at BLIULG12
SEGI - Universite de Liege
15, av. des Tilleuls
B4000 LIEGE (Belgique)
UZ32112%BLIULG12.BITNET@WISCVM.WISC.EDU
To: INFO-MICRO at BRL-VGR
Subject: Commodore 128
[line eater bait]
I am the owner of a Commodore 128 and a 1571.
I bought it for its versatility and would not mind its slow speed if
there were no exageration and it were bug free. But it is not the case.
I would like to know if these problems show up on U.S. or recent machines
or does anyone have comments or cure for the following:
128 Mode:
-The DOS shell file copy function exhibits the stange behaviour of
adding one byte to the file being copied. Harmless for program files,
it is a nuisance for text files. I'd be glad to have a correction
for this otherwise very handy program.
-The 1571 is double sided and faster than the 1541... until I get to
filling side two. On that side, writing is VERY slow. The drive
alternately accesses the data and the directory sectors for EACH
sector being written. Looks like it reads and/or writes the BAM (block
availability map) for each sector. On side one, the directory is
accessed only between a group of records. The SAVE command does however
fill both sides without directory access. It only occurs during other
I/O's (e. g. program output or file copy).
-The capslock key has no effect on the letter Q.
CP/M mode:
-Is there a way to assign the printer logical unit to the user port to
a parallel printer?
-There is a device driver supposed to drive a 6251 rs232 chip. But the
128 has none. What is it there for? I would appreciate to do rs232 I/O
(not modem).
-There is an evident lack of keyboard buffering and auto repeat.
-The screen output is very slow (80 column).
-Commodore statement of a 4MHz Z80 is a very subtle lie. It DOES 4MHz,
but only HALF of the time (each other half cycle of a 2MHz clock).
Really 2MHz. Would they pay half time employees at full rate?
-The CPM+ makes ridiculous little use of the alternate memory bank for
I/O buffers.
-In other words, is there a better CP/M than the one we get with the
machine?
[These are not opinions, but facts]UZ32112%SG2.BITNET@WISCVM.WISC.EDU (09/04/86)
Date: 4 September 1986, 15:39:06 ULG
From: Andre PIRARD +32 (41) 520180(449) UZ32112 at BLIULG12
SEGI - Universite de Liege
15, av. des Tilleuls
B4000 LIEGE (Belgique)
UZ32112%BLIULG12.BITNET@WISCVM.WISC.EDU
To: INFO-MIC at BRL
Suject: search for Forth mailing list
Does anyone know of an active Forth mailing list?
I subscribed to UMFORTH@WEIZMANN.BITNET, but they're sort of come and go.
I tried FORTH-REQUEST@SRI-CSL.ARPA but got a mail delivery reject.
Thanks.elg@usl.UUCP (Eric Lee Green) (09/05/86)
In article <3472@brl-smoke.ARPA> UZ32112%SG2.BITNET@WISCVM.WISC.EDU writes: >Date: 2 September 1986, 12:58:17 ULG >From: Andre PIRARD +32 (41) 520180(449) UZ32112 at BLIULG12 > SEGI - Universite de Liege > 15, av. des Tilleuls > B4000 LIEGE (Belgique) > UZ32112%BLIULG12.BITNET@WISCVM.WISC.EDU >To: INFO-MICRO at BRL-VGR >Subject: Commodore 128 > >[line eater bait] > >128 Mode: > >-The DOS shell file copy function exhibits the stange behaviour of >adding one byte to the file being copied. Harmless for program files, >it is a nuisance for text files. I'd be glad to have a correction >for this otherwise very handy program. Dos shell? The U.S. 128s don't come with one. There is a version of the C-64 utility "Multifilecopy" or "Copy Files" that works very well on the 128, and is excruciatingly accurate. >-The 1571 is double sided and faster than the 1541... until I get to >filling side two. On that side, writing is VERY slow. The drive >alternately accesses the data and the directory sectors for EACH >sector being written. Looks like it reads and/or writes the BAM (block >availability map) for each sector. On side one, the directory is >accessed only between a group of records. The SAVE command does however >fill both sides without directory access. It only occurs during other >I/O's (e. g. program output or file copy). Bug in the DOS. Fixed in latest release, the CBM techs assure us. What they don't assure us of, is this: When and WHERE can we get the latest release? >-The capslock key has no effect on the letter Q. Haven't noticed it. I hear that some of the very very early models of the 128 had a similiar bug where it would not print an uppercase "Q" no matter what, you might have an older model than I do. >CP/M mode: >-Is there a way to assign the printer logical unit to the user port to >a parallel printer? Sure. Get the technical manuals from Digital Research, and write a device driver. Most people, however, just buy a parallel interface that plugs into their serial port. This gives the advantage of being able to use ANY software with your printer, not just software that takes advantage of a parallel printer driver. >-There is a device driver supposed to drive a 6251 rs232 chip. But the >128 has none. What is it there for? I would appreciate to do rs232 I/O >(not modem). The latest version of the CP/M fixes this. It is available here in the U.S. from Commodore clubs and from the big time sharing services (Compuserve, The Source, Quantum Link, etc.). >-There is an evident lack of keyboard buffering and auto repeat. Fixed in latest version of CP/M. >-The screen output is very slow (80 column). Don't know if this has been fixed in the latest release. I noticed that the earlier release printed to BOTH the 40 column screen and the 80 column screen... that may have been the source of the slowness. Maybe the latest version fixes this problem. >-Commodore statement of a 4MHz Z80 is a very subtle lie. It DOES 4MHz, >but only HALF of the time (each other half cycle of a 2MHz clock). >Really 2MHz. Would they pay half time employees at full rate? True. So true. >-The CPM+ makes ridiculous little use of the alternate memory bank for >I/O buffers. Best that can be done. DR did ridiculously little to make CP/M 2.2 into CP/M 3.0. >-In other words, is there a better CP/M than the one we get with the >machine? Yes. The one you should be able to get from your local Commodore dealer or Commodore club. -- Eric Green {akgua,ut-sally}!usl!elg (Snail Mail P.O. Box 92191, Lafayette, LA 70509) " In the beginning the Universe was created. This has made a lot of people very angry and been widely regarded as a bad move."