Info-IBMPC@USC-ISIB.ARPA (Info-IBMPC Digest) (10/30/85)
Info-IBMPC Digest Tuesday, 29 October 1985 Volume 4 : Issue 126
This Week's Editor:
Today's Topics:
Epsilon 3.0 from Lugaru
SIMTEL20 Repository Status
WHEREIS.ASM Update in Library
DOSamatic Review
Problems with Hidden Files
Lotus 1-2-3
LISP/PROLOG
Hostess 8 Channel Serial Interface Query
PC ram Test Query
----------------------------------------------------------------------
Date: 26 Oct 1985 14:01:19 PDT
Subject: Epsilon 3.0 from Lugaru
From: Billy <BRACKENRIDGE@USC-ISIB.ARPA>
To: info-ibMPC@USC-ISIB.ARPA
I have been testing the latest version of Epsilon from Lugaru
Software Ltd. My beta test version is 2.99, but I am told the
official version 3.0 has been released.
Epsilon is a clone of EMACS, a popular screen editor which runs on
just about every machine and operating system worth discussing. It is
the standard editor here as it is the only editor common to
the myriad different machines in ISI's menagerie.
Epsilon now includes an extension language called EEL. (Epsilon
Extension Language) Richard Stallman, author of the original EMACS
feels strongly that extendibility is one of the most important
features of an editor and in the past has criticized the EMACS clones
on the market. I think he would be pleased with the new version of
Epsilon.
With the exception of the new extension language, Epsilon 3.0 doesn't
differ much from versions 2.xx. One major fix is in the incremental
search. It used to be difficult to use the search command in a
keyboard macro as the macro would execute whether the search was
successful or not. Adding this conditionality to the search greatly
increases the utility of keyboard macros.
Version 3.00 is also more flexible regarding the hardware
environment. One can now specify the size and location of the screen
as well as access the color attributes of various displays. Lugaru
makes the most intelligent use of search paths and environment
variables I have seen in a DOS program. Epsilon will always find its
file!
Epsilon allows one to run compiles and other well behaved programs in
background while you are editing. Background tasks only run while
Epsilon is waiting for keyboard input so it doesn't slow down editing
noticibly. I have had some problems with Texas Instruments resident
device drivers loaded while running background assemblies. I was
assured by Todd Doucet of Lugaru that Sidekick and other similar
programs work fine with this version of Epsilon. If anyone
experiences similar problems please let me or Lugaru know.
The original EMACS was written in MIT TECO, other versions have been
written in Lisp or Lisp like languages. Epsilon is written in EEL
which has the syntax of C. The entire source code for Epsilon comes on
the disk along with an EEL compiler. If there is something you don't
like about Epsilon you can change it.
I am no fan of C. I hate C. I have a closed mind on the subject.
I assume someone who likes C would love EEL. I can tolerate EEL which
coming from me is the highest praise. One of my many objections to C
is that it tends to reflect strongly the hardware architecture of the
machine it is running on. EEL is not a native compiler, but runs
on a byte coded virtual machine. The best thing about EEL is you
forget you are running on a miserable 8088. Integers are 32 bits. If
you want a new buffer call: zap(buffer) and you have a buffer that
can contain 32 bits worth of characters. You can have all the buffers
you want and have them as full as you want. Swapping isn't that bad
either. The algorithm seems to be more intelligent than your random
page fault algorithms.
Epsilon is an existence proof that well written EEL code need not be
slow. One of Epsilon's biggest selling points has been that it is
fast doing things like screen updates. As there are fast native
routines for doing most anything you might want to do with strings of
characters, you aren't spending much time in interpreted code.
One of the powerful features of the EEL machine is the regular
expression search. I am told it is more powerful than Unix's grep. I
have used it for munging messages headers in routines I wrote to
put together this digest. Once you get the hang of it is fun to
play with and I think more powerful than the sorts of things one
would have written in TECO.
Lugaru's phone service is excellent. The company is run by computer
programmers who sell a product primarily used by computer
programmers. When you call the phone is always answered by one of the
principal authors of the program (or at least they fooled me). As
I had a beta test version I probably called more often than most
people, but if they can step me through C syntax they can deal with
anybody.
Like many PC programs Epsilon is getting bigger. It requires twice
the space on the disk as the last version. This is no problem on PCs
with hard disks, but I am not throwing away my Epsilon 2.02 which is
small. I can take it travelling on a floppy disk and still have room
for other stuff.
Lugaru Software Ltd.
5740 Darlington Road
Pittsburgh, PA 15217
(412)421-5911
-------
------------------------------
Date: Sun, 27 Oct 1985 22:43 MST
From: "Frank J. Wancho" <WANCHO@SIMTEL20.ARPA>
To: INFO-CPM@AMSAA, INFO-MICRO@BRL, ADA-SW@SIMTEL20.ARPA
cc: INFO-IBMPC@USC-ISIB, INFO-HZ100@RADC-TOPS20
Subject: SIMTEL20 Repository Status
Collection Name # files # dirs ~Size(MB) # Disk Pages
MICRO:<CPM.*> 2,880 124 30.6 16,651
MICRO:<CPMUG.*> 1,923 92 12.8 6,454
MICRO:<SIGM.*> 4,795 205 40.8 21,341
MICRO:<PC-BLUE.*> 3,092 104 27.3 14,694
PS:<ADA.*> 288 28 17.3 8,176
PS:<UNIX.*> 676 26 15.8 7,743
PS:<MAIL.ARCHIVES> 101/14 lists 37.5 13,681
MICRO: is an RP06 disk drive which is full. We are anticipating the
installation of an RP07 Real Soon Now, along with a TU78 tape drive.
The RP07 has a capacity of almost three times an RP06 and has double
the channel speed. The TU78 is capable of 6250 bpi.
Shortly after the installation and check out of the RP07, all the
collections listed above, except the mail archives, will be moved to
the RP07, which will be named PD:. We will then be able to bring the
SIG/M and PC/BLUE collections up-to-date and add others as they become
available. Look for an announcement of the changeover.
Additions and changes to the CP/M collection are usually announced to
INFO-CPM. Additions to the SIG/M and PC/BLUE collections are
announced to INFO-CPM, INFO-MICRO, INFO-IBMPC, and INFO-HZ100.
Additions and changes to the Ada collection are announced only to
ADA-SW. To be added to ADA-SW, send your request to
ADA-SW-REQUEST@SIMTEL20.
Additions and changes to the Unix collection will be announced to a
new list named UNIX-SW. To be added to UNIX-SW, send your request to
UNIX-SW-REQUEST@SIMTEL20.
--Frank
------------------------------
Date: 26 Oct 1985 15:48:42 PDT
Subject: WHEREIS.ASM update added...
From: Koji Okazaki <swg.KOJI@USC-ISIB.ARPA>
To: Info-IBMPC@USC-ISIB.ARPA
cc: Koji@USC-ISIB.ARPA
The following file has been updated by Tom Brengle. Tom's source had some
problems when assembling with MASM 3.0; however, Richard Gillmann cleaned
that up. So snarfers, let's tip our hats to Tom and Richard, eh?
WHEREIS.ASM Takes a file name or pattern (e.g., *.BAS) as an
whereis.doc argument and finds where it is in the file system by
starting at the root and working down through all the
sub-directories. [Updated by Tom Brengle so that
"WHEREIS filename" does the same thing as "WHEREIS
filename.*"]
<DECVAX!MCNC!AKGUA!CLYDE!FLOYD!WHUXLE!JPH@UCB-VAX>
2/18/84
<Tom Brengle, BRENGLE%LLL@LLL-MFE> 10/23/85
-------
------------------------------
Date: Tue, 29 Oct 85 10:33:38 PST
From: walton%Deimos@CIT-Hamlet.ARPA
Subject: DOSamatic Review
I have just downloaded and attempted to use DOSamatic from UTEXAS-20. I
am not impressed, I'm afraid. First of all, I think Erik Bailey misled us
slightly, although probably unintentionally. DOSamatic is a command shell
which uses the arrow keys and pop-up menus to move around in your system,
allows single-keystroke deletion and creation of files, and so on. It is
NOT a multitasker, however. You can indeed start several programs and
switch between them using the "hot key" (ALT F10), but only the one at
which you are currently looking is active. In DoubleDOS, for example, the
background and foreground tasks operate independently. It also seems to
be relatively buggy. Here are a few:
I tried to run KERMIT as one of the tasks, and found that the Kermit
escape sequence (<control-]>C) no longer got me back to Kermit command
level. If I started an editor as a task, and jumped back to DOSamatic, the
Tasks Active indicator on the main DOSamatic screen still said 0 tasks
active. It seems to be sensitive to the placement of its necessary
environment variables--if they're too far down the list, they don't get
seen at all.
There is a general difficulty here, I think, which is that its going to
take a LOT more work, thought, and human engineering to make a command
shell or environment which is both easy to use, flexible enough for
programmers, and significantly faster and easier for a good touch typist
(which I am) than just typing the commands at the DOS prompt. DOSamatic
fulfills none of these criteria, and I don't intend to use it.
Steve Walton
Caltech Solar Astronomy
walton%deimos@cit-hamlet.arpa
walton@citdeimo.bitnet
...!psuvax1!walton@citdeimo.bitnet
------------------------------
Date: Fri, 25 Oct 85 17:30 EDT
From: "Roger C. King" <RCKing@MIT-MULTICS.ARPA>
Subject: Problems with Hidden Files
Does anyone know how to get rid of hidden files?
Roger King
------------------------------
Date: 25 Oct 1985 14:45:57 PDT
Subject: Re: Problems with Hidden Files
From: Info-IBMPC Digest <Info-IBMPC@USC-ISIB.ARPA>
You can use the CHMOD program (in the Info-IBMPC library) to
change the mode of hidden files to 0 (normal). Then delete them
in the usual way.
Dick Gillmann
------------------------------
Date: 27 Oct 1985 1829-PST
From: Rob-Kling <Kling%uci-20b.zotnet@UCI.EDU>
Subject: Lotus 1-2-3
To: burhan%ecld@usc-ecl.arpa
cc: cfmucken@ddn2.arpa, info-ibmpc@usc-isib.arpa
Jackie,
I'm about to but 7 Leading Edge PC's w/hard disks so am concerned about
potential problems of unreliability.
Given the rave reviews in PC-Week, Consumer Reports,etc. it's unlikely that
there is a basic flaw in the architecture.... however....
Here's one "elementary" guess about the problem w/Lotus 1-2-3 not
writing to drive C.
1-2-3 must set the default drive to C:
/wgdd [/worksheet, global, default, drive] must be set to C:\propersubdir.
If his Lotus 1-2-3 is configured for B: as the default drive, he can get the
problem you describe.
[BTW: I have a had a minor problem with Vterm (a terminal emulator)
generating unexpected software interrupts on the Model D. But other programs
have run smoothly.....]
I would appreciate hearing if there seems to be a design problem with
the Leading Edge Model D.
Rob Kling
UC-Irvine.
------------------------------
Date: Mon, 28 Oct 85 9:31:21 EST
From: Kenneth E. Van_Camp (LCWSL) <kvancamp@Pica-Lca.ARPA>
To: info-ibmpc@Usc-Isib.ARPA
Subject: LISP/PROLOG
To whoever was asking about LISP and PROLOG on MS/DOS systems, and the person
who subsequently noted the existence of XLISP in the public domain. I would
also like to point out that the copy of XLISP that I got from Bytenet also
included a PROLOG interpreter written in XLISP -- two birds with one stone,
if you don't need much speed and don't need the full implementation.
Also, if you don't want to fool around with C, there's a decent LISP
interpreter written in Basic by Randy Beer that appeared in 80 Micro some time
ago. It was written for TRS-80's but I easily ported it to my Sanyo 550. Let
me know if you want more info on when it appeared, or maybe I can send source
to USC-ISIB (although I don't know what copyright laws apply with regard to
80 Micro).
--Ken Van Camp (kvancamp@pica-lca.arpa)
------------------------------
From: bierma@nprdc.arpa (Larry Bierma)
Date: 29 October 1985 1329-PST (Tuesday)
To: info-ibmpc@isib
Subject: Hostess 8 channel serial interface
Greetings,
Does anyone know of a XENIX driver for the Hostess 8 channel serial
interface? Alternativly, technical documentation on the hostess board
and/or samples of other serial line drivers for XENIX would be
appreciated
--Larry ARPA: bierma@nprdc.arpa
UUCP: {decvax,ucbvax,ihnp4}!sdcsvax!sdcsla!nprdc!bierma
PSTN: (619) 225-2161
"And where are all the sportsmen who always pulled you through"
------------------------------
Date: Saturday, 24 August 1985 14:39-MDT
From: CSTROM@SIMTEL20.ARPA
Subject: PC ram test needed
ReSent-To: Info-IBMPC at USC-ISIB
ReSent-Date: Sun 27 Oct 1985 09:47-MST
Can somene point this CP/M refugee to a good public domain ram test
program for the IBM PC/XT? Such a beast _must_ exist - I am looking
for something akin to the CP/M-80 MEMR2, not the quick once-over type
of thing the rom already does. Thanks.
------------------------------
End of Info-IBMPC Digest
************************
-------