NORVELL@KNOX.BITNET (ozymandias G desiderata) (05/12/91)
First and foremost, I thank all of you who responded to my earlier query
about dealing with pernicious and ugly network files. However, I am still having
nasty problems, and I am going to restate my problem, along with a more detailed
definition of what I'm doing.
I, being the masochist that I am, am trying to retrieve the source code
for modula-2 and modula-3 from Apple's mainframe. Since I am on a VAX system
running VMS, I have to use the BITFTP server generously provided by Princeton (I
would be lost without it!). Knowing that the VAX gets very unhappy with records
longer than 512 bytes, my first command after logging on to the server is BINARY
FIXED 512. I then GET all of the files I want to look at. They are sent to me
in Netdata format. Since these files are quite chunky (+4 megs in the case of
the modula-3 distribution files), BITFTP breaks them up into bite-sized chunks
before sending them with a program called BITSEND. I have the complement
receiving script, called BITRCV, that I've modified to receive files binary-
wise instead of doing a straight receive (where the computer often attempts to
translate files from EBCDIC to ASCII. Bad in the case of binary images).
I then have a bunch of tarred and compressed files (files with .tar.Z
extensions on them (boy, all kinds of useful thingies in this letter!)). I
transfer them over to a Macintosh of my acquaintance with Kermit. I have the
public domain version for the Mac, version .98(62), and VMS version 4, or
something like that. I put both the sending and receiving kermits into binary
mode, and grab the files. The Kermit on the Mac is receiving them via Binary
mode into the Data fork of the file on this end.
Then, using MacCompress, I attempt to uncompress the files. MacCompress
is designed to work with files compressed under Unix, and it appears to be a
fairly stable program, so I don't think it's the problem. However, I get all
kinds of strange and wonderful error messages, usually resulting in a full
system crash (for the Mac), and I have to reboot.
This is frustrating me mightily, especially because of the tremendous
amounts of time it takes me to transfer the files from the VAX to the Mac (more
than a few hours because I'm receiving at a whopping 19.2 kbps. Theoretically,
I should be able to either up the serial-line speed (it's a direct connection
through a DECserver) or change the packet size to something a little bit more
manageable, like a K at a time, but the VAX doesn't like either of those in
practice).
I am wondering where the hell I'm screwing up, and what I can do to
correct this miserable situation. Any clues? I feel like a world class heel
wasting bandwidth by getting the files repeatedly, and I do truly want the
source code (that's my summer project. I want to do some ports (no, I DON'T
have a life)). I've even tried having them sent uuencoded (always a sign of
true desperation, as the files are broken down into 50k chunks. Ever tried to
concatenate 137 files, while having to strip off a variable number of header
lines from each file? I ran out of virtual memory and almost crashed the VAX
with a little program I wrote to do just that).
On a lighter note, I recently got macOberon, and I LOVE it! I love
experimental new software, especially when it's as gadget-heavy as Oberon is.
I only have one beef. How am I supposed to use it on a small (SE/30) monitor?
I cannot print any of the documentation (because of the difficulties of imple-
menting printing on printers that were not designed for the Ceres system. Ano-
ther summer project, perhaps, but maybe not, for reasons detailed below) and
I can't read large chunks of the documentation because it extends off the
screen. I know that in the paint module you can scroll horizontally, but I
haven't figured out if that's possible in Write, Edit, or Leda. Can any of you
hulk-type folks with 21" Radius TrueColors who can read the documentation help
me? Or am I simply destined to have to be spiritually impoverished until I get
a large monitor? I think I'm going to have to adopt frustration as my middle
name.
Also, is the Draw module as brain-damaged as it seems? I know,
I'm expecting a bit much, but everything else is so robust that it surprised
me that Draw was so underpowered.
Anyway, I think I've tossed you enough brainteasers for one night.
Please ask me to clarify if I've confused you. Please respond. I'm going nuts
over here, especially because it seems to me like I'm doing everything right.
I may be overlooking something minor, but I can't figure out what it is. This
time, since I have imposed on you all so much, I will personally thank each
and every one of you who replies (wow! how nifty! :-|) and will refrain from
ever sending you pictures of cows, stupid drumming jokes, or speculations on
Data's (Star Trek: The Next Generation) sexual preferences :-) Unless you really
want them.
thank you in advance,
ozymandias G desiderata
<NORVELL@KNOX.BITNET>smcgee%fuggles@CS.UTAH.EDU (Scott Mcgee) (05/14/91)
I am also interested in doing a port of Modula 2. Can you give me any
information on source code? What is this source you are trying to get?
Scott
--
The 8080 was a great chip in it's day. Pity that Intel's not been
___ been able to improve upon it. - Tim Kientzle ______________________
| Fred paid me $5 to express these opinions! They are all his! NOT mine |
|__ email: smcgee@cadesm0.eng.utah.edu or smcgee%sunset@cs.utah.edu __|