Found prints to a PDP-8P. What is it?
dbradatanu
a PDP-8P. It is all transistor, uses flip chips, and was designed by
deCatro - 1964. 12 bit.
What is it? The prints do not look like a straight-8 upon quick inspection.
...
Henry Burkhardt III
The first used Flip-Chips and the second one didn't. The second was
an un-authorized project at DEC, but Ed was concerned that the
Flip-Chip modules, on which Ken Olsen was willing to get the entire
company, would not be manufacturable, so he hired an expert circuit
designer and they designed another entire set of modules.
As it happened, Ed was right, Flip-Chips never were manufacturable
and the PDP-8 that was announced and shown in 1965 (and the one that
went into volume manufacturing) was the one that didn't use the
FlipChip modules.
The circuit designer was Richard G. Sogge. I was a programmer at DEC
at the time and wrote some simulation codes for Ed. Ed recruited me
and we recruited Dick to found Data General in 1968.
My guess is that you found prints of the orignal PDP-8 - the one that
wasn't ever built.
shsrms/aka/shsrms
>
> There were actually 2 PDP-8's designed by Edson deCastro.
>
> The first used Flip-Chips and the second one didn't. The second was
> an un-authorized project at DEC, but Ed was concerned that the
> Flip-Chip modules, on which Ken Olsen was willing to get the entire
> company, would not be manufacturable, so he hired an expert circuit
> designer and they designed another entire set of modules.
>
> As it happened, Ed was right, Flip-Chips never were manufacturable
> and the PDP-8 that was announced and shown in 1965 (and the one that
> went into volume manufacturing) was the one that didn't use the
> FlipChip modules.
>
> The circuit designer was Richard G. Sogge. I was a programmer at DEC
> at the time and wrote some simulation codes for Ed. Ed recruited me
> and we recruited Dick to found Data General in 1968.
>
> My guess is that you found prints of the orignal PDP-8 - the one that
> wasn't ever built.
And I got to support and keep all of Dick's designs running after he
left.
I just remember trying to figure out what parameters were used to choose
certain transistors with paint dots on them.....
bob
Henry Burkhardt III
<shs...@bellatlantic.net> wrote:
>And I got to support and keep all of Dick's designs running after he
>left.
>I just remember trying to figure out what parameters were used to choose
>certain transistors with paint dots on them.....
>bob
>
>
>
>
I'm a surprised to hear that!
The PDP-8 went into production in early 1966 and we didn't leave until
1968 - 2 years later. The modules with Dick's circuits should have
been fairly stable by then.
In fact, by the time we left, Ed and Dick had designed the PDP-8I
which was already in production.
Some others and I had designed the PDP-X which never went into
production!
dbradatanu
Henry Burkhardt III wrote in message
<386bf7d0...@netnews.worldnet.att.net>...
I need some clarification here. Flip Chip Modules did go into production and
were used in the classic 8 as well as the PDP-8/S. (R,S,and W,G series).
What modules were used in the PDP-8P and the PDP-X ?
Please look at http://www.pdp8.com for some pictures.
>production!
Henry Burkhardt III
and those models were designed by Dick Sogge (and some others) in 1964
and early 1965. They used discrete transistors, diodes, caps and
resistors.
DEC had been counted on FlipChips which were diodes, caps and thick
films resistors fabricated on a ceramic base. This would have reduced
module cost substantially and a "standard line" of DEC modules were
designed around them. The original PDP8 design used these FlipChip
modules. It turned out that DEC couldn't manufacture them, except in
laboratory quantities.
Ed had been afraid of this happening, so, in a project not authorized
by Ken, he hired Dick and they designed an alternative set of modules.
They were not pin compatable with the FlipChip modules, so there was a
different PDP8 design.
Have we met in the 1960s?
regards,
Henry
shsrms/aka/shsrms
their eyes on a new machine, and they did not have time to support all
the 8 boards, making sure the transistors were properly matched..
I was kind of fanatical about making sure everything worked exactly
right.
When Murry Rubin (KV8i et al) left to help start DTS, I got to work with
Al Deluca
on the KV8e, but that kind of ....
Now what were the dimensions on the X? The board size....
If my memory serves right, wasn't a good part of the discussion about
whether or not
the quad flipchips would withstand the wave solder machine?
I am rusty here, but the system modules, what were the boards made
from? I seem to
recall some kind of pressed paper like stuff we calle guano....like in
the vt50 design that Stockebrand led.
It would have been great to work with you guys!!
Thanks for some great designs!
bob
Henry Burkhardt III wrote:
>
> On Thu, 30 Dec 1999 03:10:07 GMT, shsrms/aka/shsrms
> <shs...@bellatlantic.net> wrote:
>
> >And I got to support and keep all of Dick's designs running after he
> >left.
> >I just remember trying to figure out what parameters were used to choose
> >certain transistors with paint dots on them.....
> >bob
> >
> >
> >
> >
>
> I'm a surprised to hear that!
>
> The PDP-8 went into production in early 1966 and we didn't leave until
> 1968 - 2 years later. The modules with Dick's circuits should have
> been fairly stable by then.
>
> In fact, by the time we left, Ed and Dick had designed the PDP-8I
> which was already in production.
>
> Some others and I had designed the PDP-X which never went into
> production!
David M. Razler
| There were actually 2 PDP-8's designed by Edson deCastro.
|
| The first used Flip-Chips and the second one didn't. The second was
| an un-authorized project at DEC, but Ed was concerned that the
| Flip-Chip modules, on which Ken Olsen was willing to get the entire
| company, would not be manufacturable, so he hired an expert circuit
| designer and they designed another entire set of modules.
|
| As it happened, Ed was right, Flip-Chips never were manufacturable
| and the PDP-8 that was announced and shown in 1965 (and the one that
| went into volume manufacturing) was the one that didn't use the
| FlipChip modules.
|
| The circuit designer was Richard G. Sogge. I was a programmer at DEC
| at the time and wrote some simulation codes for Ed. Ed recruited me
| and we recruited Dick to found Data General in 1968.
|
Er, every DEC machine from the PDP-7 to the PDP-15 (excluding the later 8s and
11s) was built out of what I always thought were Flip Chips (That's what all
the RWBGAKM and S-series* modules *say* they are, anyway)
The RBS and early WG and A series were discrete-transistor modules replacing
the much bulkier SMBs (system module blocks), those aluminum-framed things
with big heavy connectors and lots of real estate
The later W,G,A and all the M & K Flip Chips were SSI for the most part.
So what were Flip Chips (the ones Ken bet on) *supposed* to be?
dmr
*The SMBs were sold in three speeds (standard in Hz) 2000,4000,6000 - and I
believe most, if not all were germanium-based
The Flip Chip series were almost all named for the color of their handle
(which designated their purpose, but without any particular reason)
R ed- general purpose medium speed discrete silicon
S pecial - Red handled red modules capable of additional fan-out
B lue - same as Red but faster
W hite - Digital I/O, connectors, bus and lamp drivers
G reen - Core memory and power
A mber - A/D D/A and support circutry for same
M agenta - DEC goes to SSI and positive logic, matching W,G,A modules produced
blac K - M modules for industrial environment
The early general purpose modules in all series had the letter and a three
digit code stamped on the handle, and most of the time you could guess the
function from the number.
X-2XX modules, for instance, were almost always flipflops, X-4XX modules were
usually clocks, etc.
David M. Razler
david....@worldnet.att.net
HerbalGypsy/justbob
Don't forget, the boards before the "flip chips" were called system
modules,
I don't recall the composition of the materials, but something like
pressed
paper is what I recall, or some sort of reprocessed guano like in the
base of
the VT50, with an aluminum frame.
There was some concern about the wave soldering machines, the thickness
of the
boards, how the boards would stand up to wave solder, the quality of
wave soldering,
and don't forget the time table.....
The flip chips that most folks recall as the single or dual finger set
boards
were a break through - with wave soldering technique making them mass
producable
vice the benchtop soldering by the ladies in building 5.
The big revolution, to get more space for components, on a glass base,
was -
from my recollection - just one component of the impetus that had HB,
ED, and RS
et al move on to form DG.
There was more to the story of course, and the thought of a more
powerful 8, say a
16 bit 8, etc.
Creative spirits, great ideas, challenging times, and folks with very
clear
visions of where things should go all led to this computer revolution!!
The 8/e was the first Quad board 8. using extended 4 flip chip panels.
The KI, KL, 11, etc with the larger size boards still carried the label
flip chip!
bob
dbradatanu
David M. Razler wrote in message ...
>h...@alumni.Princeton.edu (Henry Burkhardt III) wrote:
>
<snip>
>
>So what were Flip Chips (the ones Ken bet on) *supposed* to be?
It's quite confusing. I have quite a bit of documentation here that refers
to things DEC initally named one thing that we call another now. From what I
can tell, DEC actually considered a true "Flip Chip" to be a board with
resistors, diodes, and capacitors fabricated on a ceramic base/board. DEC
was unable to mass produce these boards and ended up producing what we call
"Flip Chips" now.
BTW: Now , I have a question about the PDP-11/20. I have read on many
websites that the PDP-11/20 used "M-series" flip chips. I have 8 different
PDP-11/20s... all with serial numbers from 229 to 1600. None of them use
M-series single width flip chips. In fact, they all use quad "M series Flip
Chip" boards. I haven't opened them all up yet but it looks like they are
all going to have quad width boards (according to the microfiche and
prints).
Did DEC ever produce a PDP-11 with single width M-series boards?
HerbalGypsy/justbob
and 11/15.
The msel boards, I don't recall the numbers, the ir boards I think,
there were a couple of standard boards used before the circuitry was
shrunk and before the hex boards came out.
bob
John Everett
david....@worldnet.att.net says...
>
>Er, every DEC machine from the PDP-7 to the PDP-15 (excluding the later 8s
and
>11s) was built out of what I always thought were Flip Chips
I recall at least one person in the mill had a bag of "Flip Chip" doggie
treats thumb-tacked to the outside of their cubicle. Does anyone else remember
these?
Also reminds me of the full page ad for the VAX, a British made vacuum cleaner
that someone had displayed on the wall of their cube...I believe in Marlboro.
--
jeverett<AT>wwa<DOT>com (John Everett) http://www.wwa.com/~jeverett
HerbalGypsy/justbob
bob
Douglas W. Jones,201H MLH,3193350740,3193382879
If you look at early DEC handbooks (1965, 1966), you find that the flip
chip is a hybrid integrated circuit technology, using thin-film ceramic
substrates with silicon dice bonded to the substrate for components such
as diodes, and resistors directly constructed from the circuit traces on
the substrate.
These things only showed up on the S series of what DEC came to call
Flip Chip Modules; by the time the M series came out, the term Flip Chip
was indellibly attached to the circuit board format, independent of the
use of Flip Chip hybrid integrated circuits.
The curious thing is that Flip Chip technology is still with us! The
term comes from the fact that the silicon chip is flipped over and bonded
face-down to the ceramic substrate, with solder bumps on the contact
pads of the substrate aligned with the contact pads on the chip and then
ultrasonically bonded.
Nowdays, if you do a web search on the term Flip Chip, you'll get plenty
of hits from sites related to this modern manufacturing technique, a
technique that DEC pioneered the use of with the R and S series of modules.
Doug Jones
jo...@cs.uiowa.edu
dbradatanu
about 4 quad width boards. I just checked two more 11/20s and they have the
same config. I also checked two 11/15s I have here and the same thing: Some
quad width, several dual width, many single width.. They all have the same
config.
I was hoping that DEC made the 11/20s with the same M-series modules as the
8L but then again.. the mini would have been very large.
Oh well.....
--
PDP 8s and other minis...
HerbalGypsy/justbob wrote in message <388DC7D8...@bellatlantic.net>...
>Yes. Extended, single width boards were used in the beginning with 11/20
>and 11/15.
>The msel boards, I don't recall the numbers, the ir boards I think,
>there were a couple of standard boards used before the circuitry was
>shrunk and before the hex boards came out.
>bob
>
>
>dbradatanu wrote:
>>
>> David M. Razler wrote in message ...
>> >h...@alumni.Princeton.edu (Henry Burkhardt III) wrote:
>> >
>>
>> <snip>
>>
>> >
>> >So what were Flip Chips (the ones Ken bet on) *supposed* to be?
>>
Roger Ivie
>> So what were Flip Chips (the ones Ken bet on) *supposed* to be?
>
>chip is a hybrid integrated circuit technology, using thin-film ceramic
>substrates with silicon dice bonded to the substrate for components such
>as diodes, and resistors directly constructed from the circuit traces on
>the substrate.
>
>term comes from the fact that the silicon chip is flipped over and bonded
>face-down to the ceramic substrate, with solder bumps on the contact
>pads of the substrate aligned with the contact pads on the chip and then
>ultrasonically bonded.
There's a good discussion of flip-chip technology in the book "IBM's 360 and
early 370 computers." I forget the author. The book is excellent; I couldn't
put it down (I did skip the chapter on software, though).
--
Roger Ivie
TeraGlobal Communications Corporation
1770 North Research Park Way Suite 100
Logan, UT 84341
mailto:ri...@teraglobal.com
phoneto:(435)787-0555
faxto:(435)787-0516
-----------== Posted via Newsfeeds.Com, Uncensored Usenet News ==----------
http://www.newsfeeds.com The Largest Usenet Servers in the World!
------== Over 73,000 Newsgroups - Including Dedicated Binaries Servers ==-----
Douglas W. Jones,201H MLH,3193350740,3193382879
by "dbradatanu" <dan...@internet.look.ca>:
> The 11/20s I have here have quite a few extended single width boards with
> about 4 quad width boards. I just checked two more 11/20s and they have the
> same config. I also checked two 11/15s I have here and the same thing: Some
> quad width, several dual width, many single width.. They all have the same
> config.
>
> I was hoping that DEC made the 11/20s with the same M-series modules as the
> 8L but then again.. the mini would have been very large.
By the time DEC was making the 11/15 and 11/20 (the 11/15 was the OEM
version of the /20, just as the 8/M was the OEM version of the 8/F),
they had sufficient support for board-level design that it was quite
cost-effective to make custom boards for CPUs. The result was that no
general purpose M-series boards showed up in any new CPU designs. On
the other hand, the new boards designed for the PDP-11 and the PDP-8/E
(designed concurrently and released about the same time) were all given
M-series numbers, and general purpose M-series boards still found their
way into some peripherals, particularly, those made in low enough volume
that custom wire-wrap with mostly standard boards was a less expensive
way to go than custom boards.
Even the original -8 had a number of boards that were unique to that
machine, and not considered part of the general purpose flip-chip
design, and this continued to be true in the 8/I and 8/L with the move
to M-series logic, but these were mixed liberally with standard boards.
Occasionally, some of these special purpose boards were pushed into
the standard product line, particularly things like bus decoders that
were useful components in homebrew peripherals and whatnot.
As DEC's CAD support got better, only a few years later, the use of
"standard" M-series logic was almost extinguished except for maintenance
of older systems and one-off hand-built prototypes.
Doug Jones
jo...@cs.uiowa.edu
Ric Werme
>I recall at least one person in the mill had a bag of "Flip Chip" doggie
>treats thumb-tacked to the outside of their cubicle. Does anyone else remember
>these?
I remember an cubicle with a bag of "Flip Chip" doggie treats thumb-tacked
to the outside. I believe Hartz still sells them, but not having a dog, I
don't keep up with such info.
>Also reminds me of the full page ad for the VAX, a British made vacuum cleaner
>that someone had displayed on the wall of their cube...I believe in Marlboro.
That may have been after my time, but I heard about "Nothing sucks
like a VAX."
At CMU we discovered that the TU-10 tape drive used vacuum motors that had
the same form factor as some household vacuum cleaner. Those motors were
a lot cheaper but didn't last as long, were louder, and sounded like a
vacuum cleaner. We used them anyway. Certainly didn't harm the TU-10's
performance (which sucked).
How that for subject drift?
--
Ric Werme | we...@nospam.mediaone.net
http://people.ne.mediaone.net/werme | ^^^^^^^ delete
Peter da Silva
Ric Werme <we...@nospam.mediaone.net> wrote:
> That may have been after my time, but I heard about "Nothing sucks
> like a VAX."
Mixing up two ad campaigns. It was "Nothing sucks like Electrolux".
--
In hoc signo hack, Peter da Silva <pe...@baileynm.com>
`-_-' Ar rug tú barróg ar do mhactíre inniu?
'U`
"I *am* $PHB" -- Skud.
Ambrose, Joseph
a similar font style as I recall.....
I wonder why DEC didn't sic their lawyers on that one ( he he.... )
--
Take out the extra dash in my email address if you wish to send mail.
--
Joseph Ambrose
NT Network Administrator /
Open VMS System Manager
The Conference Board
joe.a...@conference--board.org
ICQ# 13652219
John Everett <jeve...@wwa.DEFEAT.UCE.BOTS.com> wrote in message
news:j%jj4.1739$t_2....@ord-read.news.verio.net...
> In article <ldop8s8kis8l5i2cp...@4ax.com>,
> david....@worldnet.att.net says...
> >
> >Er, every DEC machine from the PDP-7 to the PDP-15 (excluding the later
8s
> and
> >11s) was built out of what I always thought were Flip Chips
>
> I recall at least one person in the mill had a bag of "Flip Chip" doggie
> treats thumb-tacked to the outside of their cubicle. Does anyone else
remember
> these?
>
> Also reminds me of the full page ad for the VAX, a British made vacuum
cleaner
> that someone had displayed on the wall of their cube...I believe in
Marlboro.
>
Mike Meredith at home
In article <cftk4.2576$Hz5....@news.optonline.net>,
"Ambrose, Joseph" <joe.a...@conference-board.org> writes:
> I remember seeing the ad for the VAX vacuum cleaner. I think the name was in
> a similar font style as I recall.....
Well, it wasn't an uncommon font style.
>
> I wonder why DEC didn't sic their lawyers on that one ( he he.... )
Perhaps the VAX vacuum cleaner was older ?
According to rumour, the two relevant presidents met and decided
that the chances of anybody confusing the two VAXes was very low
and decided to keep the lawyers out of it.
Antonio Carlini
>
>I remember seeing the ad for the VAX vacuum cleaner. I think the name was in
>a similar font style as I recall.....
>
Because the VAX vacuum cleaner (which is still sold to this day BTW) was there
first. If anyone could have sued anyone then DEC would have been on the
receiving end.
Antonio
Antonio Carlini Mail: car...@true.lkg.dec.com
DECnet-Plus for OpenVMS Engineering
COMPAQ Reading, UK
Marty
<we...@nospam.mediaone.net> wrote:
>jeve...@wwa.DEFEAT.UCE.BOTS.com (John Everett) writes:
>
>>I recall at least one person in the mill had a bag of "Flip Chip" doggie
>>treats thumb-tacked to the outside of their cubicle. Does anyone else remember
>>these?
>
>I remember an cubicle with a bag of "Flip Chip" doggie treats thumb-tacked
>to the outside. I believe Hartz still sells them, but not having a dog, I
>don't keep up with such info.
>
>>Also reminds me of the full page ad for the VAX, a British made vacuum cleaner
>>that someone had displayed on the wall of their cube...I believe in Marlboro.
>
>That may have been after my time, but I heard about "Nothing sucks
>like a VAX."
>
>At CMU we discovered that the TU-10 tape drive used vacuum motors that had
>the same form factor as some household vacuum cleaner. Those motors were
>a lot cheaper but didn't last as long, were louder, and sounded like a
>vacuum cleaner. We used them anyway. Certainly didn't harm the TU-10's
>performance (which sucked).
>
>How that for subject drift?
>--
>Ric Werme | we...@nospam.mediaone.net
>http://people.ne.mediaone.net/werme | ^^^^^^^ delete
The problem was the TU10 vac had no real good way to attach the bag ;)
And as for sound.. nothing has a sound like that of the tape as it
falls off of the reel when the sensors fails in the tape vacum
columns..."ferrrrrrrrrreeeepppp!!...bllluuuuuhhh!". Oh well so much
for the days when you had something to adjust rather than just
replace.
Ira, 72J
Alan H. Martin
>
> In article <bWqj4.42928$zU5.5...@wbnws01.ne.mediaone.net>,
> Ric Werme <we...@nospam.mediaone.net> wrote:
> > like a VAX."
>
That flies in the face of the chronology:
http://www.tuxedo.org/~esr/jargon/html/entry/VAX.html
Late 60's: Electrolux vacuum ad campaigns.
1978: VAX release.
1986-1992: VAX vacuum ad campaigns.
I doubt the computer slogan predates the UK VAX vacuum ad campaign.
/AHM
--
Alan Howard Martin AMa...@MA.UltraNet.Com
Megan
>Hey, I remain pissed off at the sanitizing of PDP-10 references from
>the file. But he doesn't make any personal claims about the chronology,
>just relates what others have told him.
I'm with you... I'll never forgive him for what he has done to the
file.
Any idea where I can find a pre-munged (is *mung* still there?)
copy of it?
Megan Gentry
Former RT-11 Developer
+--------------------------------+-------------------------------------+
| Megan Gentry, EMT/B, PP-ASEL | Internet (work): gentry!zk3.dec.com |
| Unix Support Engineering Group | (home): mbg!world.std.com |
| Compaq Computer Corporation | addresses need '@' in place of '!' |
| 110 Spitbrook Rd. ZK03-2/T43 | URL: http://world.std.com/~mbg/ |
| Nashua, NH 03062 | "pdp-11 programmer - some assembler |
| (603) 884 1055 | required." - mbg |
+--------------------------------+-------------------------------------+
Tim Shoppa
>
> "Alan H. Martin" <AMa...@MA.UltraNet.Com> writes:
>
> >Hey, I remain pissed off at the sanitizing of PDP-10 references from
> >the file. But he doesn't make any personal claims about the chronology,
> >just relates what others have told him.
>
> I'm with you... I'll never forgive him for what he has done to the
> file.
I'm not so sure the changes were all for the worse. Sure, there
were PDP-10 things excised that I would've left in, but the editor has
to draw the line between "arcana" and "jargon" somewhere. And some
of the entries have gone from "computer-hacker jargon" to "mainstream
slang" in the meantime, too.
> Any idea where I can find a pre-munged (is *mung* still there?)
> copy of it?
There's a version from 1981 or so in the [374,1] directory of the
Fall 1981 RSX-11 SIGtape. If you don't have all the RSX sigtapes
handy there, Megan, you can find this particular file over the web
at
http://www.trailing-edge.com/www/jargon.txt
If you look at Joe Smith's archive of alt.sys.pdp10 stuff at
http://www.inwap.com/pdp10/usenet/jargon
you'll find some of the differences between the "old jargon" and
"new jargon" entries discussed. The actual separation out of
the "old" PDP-10 entries was discussed at great length in the
very early 90's on alt.folklore.computers; if anyone knows where
these discussions may have been archived, I'd be very interested
in snarfing copies.
--
Tim Shoppa Email: sho...@trailing-edge.com
Trailing Edge Technology WWW: http://www.trailing-edge.com/
7328 Bradley Blvd Voice: 301-767-5917
Bethesda, MD, USA 20817 Fax: 301-767-5927
Peter da Silva
Alan H. Martin <AMa...@MA.UltraNet.Com> wrote:
> > Eric's Jargon File work is laudable, but he's just another hacker, and he's
> > been mistaken before.
>
> Hey, I remain pissed off at the sanitizing of PDP-10 references from
> the file.
I'm not pissed off at Eric, I'm just aware that he tends to be a bit willing
to accept stuff as gospel that ain't necessarily so.
Richard M. Alderson III
TOAD:<INFO>
JARGON.TXT.1;P775252 34 85768(7) 23-Jul-1981 08:04:38 HELLIWELL
Total of 34 pages in 1 file
This version, also dated 1981, is available via anonymous FTP to toad.xkl.com.
Rich Alderson Last LOTS Tops-20 Systems Programmer, 1984-1991
Current maintainer, MIT TECO EMACS (v. 170)
Customer Advocacy, XKL LLC, 1993-now
last name @ XKL dot COM Chief systems administrator, XKL LLC, 1998-now
Johnny Billquist
> >just relates what others have told him.
>
> I'm with you... I'll never forgive him for what he has done to the
> file.
Hear hear. And I thought I was the only one to dislike Raymonds
munging. (Hmmm, I think I heard that MRC didn't like it either,
but I'm not sure...)
> Any idea where I can find a pre-munged (is *mung* still there?)
> copy of it?
Should be available here and there, I think. If nothing else helpes,
I guess I could go hunting our PS: for it.
(Well, a copy on an emulator anyway, since the real 2060 is not bootable
right now).
Johnny
--
Johnny Billquist | johnny.b...@netinsight.net
Net Insight AB | phone: +46 8 685 04 88
Västberga Allé 9 | fax: +46 8 685 04 20
Box 42093 |
SE-126 30 STOCKHOLM, Sweden | http://www.netinsight.net
Johnny Billquist
>
> In article <38AEB313...@ma.ultranet.com>,
> Alan H. Martin <AMa...@MA.UltraNet.Com> wrote:
> > > Eric's Jargon File work is laudable, but he's just another hacker, and he's
> > > been mistaken before.
> >
> > the file.
>
> to accept stuff as gospel that ain't necessarily so.
I'm pretty pissed. It's not only that he accept anything as gospel.
When he started munging JARGON.TXT he was really creative in adding
his own stuff, removing things he didn't understand, and totally
missing the point of the document. It turned into just another
piece of junk to be used by the same people who like to call themself
hackers and all that stuff.
Peter da Silva
Johnny Billquist <Johnny.B...@netinsight.se> wrote:
>I'm pretty pissed. It's not only that he accept anything as gospel.
>When he started munging JARGON.TXT he was really creative in adding
>his own stuff, removing things he didn't understand, and totally
>missing the point of the document. It turned into just another
>piece of junk to be used by the same people who like to call themself
>hackers and all that stuff.
The original document still exists. There's no reason you can't promote
it yourself... maybe even update it with sufficient commentary that people
in today's Windowscentric world can understand it.
This isn't like colorizing a movie where the original movie quits being
broadcast, published on tape and disk, and screened. It's on the net, you
can keep it alive without megabucks.
--
This is The Reverend Peter da Silva's Boring Sig File - there are no references
to Wolves, Kibo, Discordianism, or The Church of the Subgenius in this document
Executive Vice President, Corporate Communications, Entropy Gradient Reversals.
Tim Shoppa
> > >just relates what others have told him.
> >
> > I'm with you... I'll never forgive him for what he has done to the
> > file.
>
> Hear hear. And I thought I was the only one to dislike Raymonds
> munging. (Hmmm, I think I heard that MRC didn't like it either,
> but I'm not sure...)
I don't think that what ESR removed was all that drastic.
Yes, he excised a few PDP-10 specific references - but *not*
very many. By my count (see my postings from a few months ago
on this subject, or the copy at Joe Smith's archive at
http://www.inwap.com/pdp10/usenet/jargon
) only 8 entries disappeared. Yes, some of those 8 entries have
historical importance, and *I* would've left them in myself, but
it's not so awful that they were removed.
> > Any idea where I can find a pre-munged (is *mung* still there?)
> > copy of it?
>
> Should be available here and there, I think. If nothing else helpes,
> I guess I could go hunting our PS: for it.
It's there on the RSX CD-ROM archives I sent you, Johnny. The
[374,1] directory of the Fall 1981 RSX SIG tape. It's also available
over the web (didn't I just say this a day or two ago?) at
http://www.trailing-edge.com/www/jargon.txt
Tim Shoppa
>
> Some people were annoyed by his changing references that denigrated UNIX
> to refer instead to MS-DOS, since, as we all know, UNIX is what hackers love.
It depends on the hacker :-). I, for one, am a proud owner of
a copy of _The Unix-Hater's Handbook_ by Garfinkel, Weise, and
Strassmann. It's not a perfect book - many of the contributors
were so enraged by Unix's superficial design faults that they
never got around to discussing its fundamental design faults -
but it's a very worthwhile and enjoyable resource.
What's truly sad is that there's a whole new generation - called
"slashdotters" after the weenie web site where they hang out at -
that have never seen an OS better than Linux. To quote
Ken Pier of Xerox PARC:
I liken starting one's computing career with Unix, say as
an undergraduate, to being born in East Africa. It is
intolerably hot, your body is covered with lice and flies, you
are malnourished and you suffer from numerous curable diseases.
But, as far as young East Africans can tell, this is simply the
natural condition and they live within it. By the time they
find out differently, it is too late. They already think
that the writing of shell scripts is a natural act.
Tim.
Christopher C Stacy
Jay Maynard
wrote:
>Some people were annoyed by his changing references that denigrated UNIX
>to refer instead to MS-DOS, since, as we all know, UNIX is what hackers love.
Ah, I see.
All of the complaining is from embittered old ITS hackers pissed off that
their beloved OS has sunk beneath the waves.
<music href="mp3://Eagles/Get Over It!">
I miss OS/2, too, but I don't dedicate a chunk of my brainpower to it any
more...there are too many other interesting systems to play with. Like ITS
that much? Build a PDP-10 emulator and RUN IT!
</music>
ESR's changing of those refs seems to me to make the entries in question
more universally understandable. Lots of folks like Unix, whether you choose
to admit and understand it or not. Many, many fewer like MS-DOS.
Peter Seebach
Tim Shoppa <sho...@trailing-edge.com> wrote:
>It depends on the hacker :-). I, for one, am a proud owner of
>a copy of _The Unix-Hater's Handbook_ by Garfinkel, Weise, and
>Strassmann. It's not a perfect book - many of the contributors
>were so enraged by Unix's superficial design faults that they
>never got around to discussing its fundamental design faults -
>but it's a very worthwhile and enjoyable resource.
FWIW, a lot of the "Unix" flaws these people talk about are either
flaws in programming languages, or simply *misunderstandings* of Unix
interfaces.
Unix is an awful MS-DOS clone, and a fairly poor Mac. It's a pretty
good Unix.
>What's truly sad is that there's a whole new generation - called
>"slashdotters" after the weenie web site where they hang out at -
>that have never seen an OS better than Linux.
Uh-huh. As a BSD bigot, I'd rather see Linux in that position than Windows,
but it's still frustrating.
-s
--
Copyright 2000, All rights reserved. Peter Seebach / se...@plethora.net
C/Unix wizard, Pro-commerce radical, Spam fighter. Boycott Spamazon!
Consulting & Computers: http://www.plethora.net/
Get paid to surf! No spam. http://www.alladvantage.com/go.asp?refid=GZX636
Johnny Billquist
>
> Johnny Billquist wrote:
> >
> > Megan wrote:
> > >
> > > "Alan H. Martin" <AMa...@MA.UltraNet.Com> writes:
> > >
> > > >Hey, I remain pissed off at the sanitizing of PDP-10 references from
> > > >the file. But he doesn't make any personal claims about the chronology,
> > > >just relates what others have told him.
> > >
> > > I'm with you... I'll never forgive him for what he has done to the
> > > file.
> >
> > Hear hear. And I thought I was the only one to dislike Raymonds
> > munging. (Hmmm, I think I heard that MRC didn't like it either,
> > but I'm not sure...)
>
> I don't think that what ESR removed was all that drastic.
> Yes, he excised a few PDP-10 specific references - but *not*
> very many. By my count (see my postings from a few months ago
> on this subject, or the copy at Joe Smith's archive at
>
> http://www.inwap.com/pdp10/usenet/jargon
>
> ) only 8 entries disappeared. Yes, some of those 8 entries have
> historical importance, and *I* would've left them in myself, but
> it's not so awful that they were removed.
Hmmm. I seem to remmeber, from browsing the files, that I easily
could find more than eight removed entries... Maybe I should look
again. (Or maybe I just overreactive from all the new crud that
got stuffed in there.)
> > > Any idea where I can find a pre-munged (is *mung* still there?)
> > > copy of it?
> >
> > Should be available here and there, I think. If nothing else helpes,
> > I guess I could go hunting our PS: for it.
>
> It's there on the RSX CD-ROM archives I sent you, Johnny. The
> [374,1] directory of the Fall 1981 RSX SIG tape. It's also available
> over the web (didn't I just say this a day or two ago?) at
>
> http://www.trailing-edge.com/www/jargon.txt
You did say it. Don't mean I didn't post before your information though. :-)
(And JARGON.TXT wasn't the first thing I started looking for on the
RSX freeware CD. Excellent item by the way.)
Johnny Billquist
>
> Christopher C Stacy wrote:
> >
> > Some people were annoyed by his changing references that denigrated UNIX
> > to refer instead to MS-DOS, since, as we all know, UNIX is what hackers love.
I don't know which version of the JARGON.TXT you've been reading (probably
jargon.txt instead), Cristopher.
JARGON.TXT whacks down on Unix pretty hard. Anyone who think JARGON.TXT
is advocating Unix should know that this is also something that Raymond
did, along with adding MS-DOS stuff.
> It depends on the hacker :-). I, for one, am a proud owner of
> a copy of _The Unix-Hater's Handbook_ by Garfinkel, Weise, and
> Strassmann. It's not a perfect book - many of the contributors
> were so enraged by Unix's superficial design faults that they
> never got around to discussing its fundamental design faults -
> but it's a very worthwhile and enjoyable resource.
I really guess I should read it sometime. I think I would
ejoy it.
> What's truly sad is that there's a whole new generation - called
> "slashdotters" after the weenie web site where they hang out at -
> that have never seen an OS better than Linux. To quote
> Ken Pier of Xerox PARC:
>
> I liken starting one's computing career with Unix, say as
> an undergraduate, to being born in East Africa. It is
> intolerably hot, your body is covered with lice and flies, you
> are malnourished and you suffer from numerous curable diseases.
> But, as far as young East Africans can tell, this is simply the
> natural condition and they live within it. By the time they
> find out differently, it is too late. They already think
> that the writing of shell scripts is a natural act.
Yup. This is what I'm complaining about...
Thanks for an excellent name for them. Never heard it before.
"slashdotters". I like name calling. :-)
Christopher C Stacy
Jay> On Tue, 22 Feb 2000 01:28:22 GMT, Christopher C Stacy <cst...@world.std.com>
Jay> wrote:
>> Some people were annoyed by his changing references that denigrated UNIX
>> to refer instead to MS-DOS, since, as we all know, UNIX is what hackers love.
Jay> Ah, I see.
Jay> All of the complaining is from embittered old ITS hackers pissed off that
Jay> their beloved OS has sunk beneath the waves.
I made a remark about string replacing "UNIX" with "MS-DOS",
and am unable to comprehend your bizarre message about ITS.
Christopher C Stacy
ITS was rather specific to the PDP-10, and never ran on the same
hardware that UNIX or MS-DOS run on. (However, you can run a PDP-10
emulator on any of those things now, and run ITS on top of that.
It's been run on SUN boxes and Macintosh laptops and all kinds of things.)
Christopher C Stacy
Johnny> I don't know which version of the JARGON.TXT you've been reading (probably
Johnny> jargon.txt instead), Cristopher.
Johnny> JARGON.TXT whacks down on Unix pretty hard. Anyone who think JARGON.TXT
Johnny> is advocating Unix should know that this is also something that Raymond
Johnny> did, along with adding MS-DOS stuff.
Uh, that's what I said?
Johnny Billquist
Oh? I thought you said that hackers love Unix, and that Unix got replaced with
MS-DOS in JARGON.TXT.
Mark Crispin
> Hear hear. And I thought I was the only one to dislike Raymonds
> munging. (Hmmm, I think I heard that MRC didn't like it either,
> but I'm not sure...)
MRC detested it, and did his best to prevent the developing travesty.
But the world was filled with ravening hordes of UNIX barbarians, who were
determined to sack and pillage what was left of the PDP-10 world and leave
no traces of machines that didn't think in 8-bit bytes.
The position proved hopeless, and MRC was forced to capitulate.
-- Mark --
* RCW 19.190 notice: This email address is located in Washington State. *
* Unsolicited commercial email may be billed $500 per message. *
Science does not emerge from voting, party politics, or public debate.
Jay Maynard
wrote:
> Jay> Ah, I see.
> Jay> All of the complaining is from embittered old ITS hackers pissed off that
> Jay> their beloved OS has sunk beneath the waves.
>I made a remark about string replacing "UNIX" with "MS-DOS",
>and am unable to comprehend your bizarre message about ITS.
You mean there are Unix haters who never touched ITS? Amazing. I'd thought
that that was the font of all that vitriol. (Unless you're one of the legion
of NT admins who think that any OS that doesn't come from Redmond
automatically sucks, but I doubt that strongly in your case. You seem too
intelligent for that.)
>ITS was rather specific to the PDP-10, and never ran on the same
>hardware that UNIX or MS-DOS run on. (However, you can run a PDP-10
>emulator on any of those things now, and run ITS on top of that.
>It's been run on SUN boxes and Macintosh laptops and all kinds of things.)
I hadn't heard that anyone had ITS running on a -10 emulator...
(To forestall an argument: No, I don't think Unix is the be-all and end-all
of OSes, but I do believe it's a superior programmer's environment for lots
of things, and there are more than a few concepts that Unix got right and
other OSes designed today still get wrong. Even so, there are places Unix is
not best suited. (I say that as I run yet another OS/360 gen on another
machine...))
Peter da Silva
Johnny Billquist <Johnny.B...@netinsight.se> wrote:
> Oh? I thought you said that hackers love Unix, and that Unix got replaced with
> MS-DOS in JARGON.TXT.
I think your sarcasm module is overdue for its 50,000 gripe checkup.
Paul Grayson
> "slashdotters" after the weenie web site where they hang out at -
> that have never seen an OS better than Linux. To quote
> Ken Pier of Xerox PARC:
>
I stopped reading Slashdot when I realised the mentality of some of the
posters. I would sometimes post a response to a thread, usually to consider
a different techology. I'd often receive a response to the effect "Are you
on crack!", or even worse. I sometimes would receive abusive emails from
people I'd angered. Sometimes it is not much different to the letters pages
of early 80s micro magazines. And the majority are ignorant of computer
history, too!
I am a firm believer in using the best technology for the job, no matter who
produced it. These views don't often go down well on Slashdot.
Eric Fischer
> Some people were annoyed by his changing references that denigrated UNIX
> to refer instead to MS-DOS, since, as we all know, UNIX is what hackers love.
I've just taken a look through JARGON.TXT (version of July 22, 1981)
and can't find a single reference, positive or negative, to either Unix
*or* MS-DOS. There are, however, a few disparaging comments made about
Multics. I have certain disagreements with Eric Raymond myself, but I
can't see the basis for this particular claim.
eric
Christopher C Stacy
Jay> I hadn't heard that anyone had ITS running on a -10 emulator...
I think that ITS was the first thing to run on the first PDP-10
emulator (KLH-10), quite a few years ago.
Christopher C Stacy
Johnny> Christopher C Stacy wrote:
>>
>> >>>>> On Tue, 22 Feb 2000 07:53:57 GMT, Johnny Billquist ("Johnny") writes:
Johnny> I don't know which version of the JARGON.TXT you've been reading (probably
Johnny> jargon.txt instead), Cristopher.
Johnny> JARGON.TXT whacks down on Unix pretty hard. Anyone who think JARGON.TXT
Johnny> is advocating Unix should know that this is also something that Raymond
Johnny> did, along with adding MS-DOS stuff.
>>
>> Uh, that's what I said?
Johnny> Oh? I thought you said that hackers love Unix, and that Unix got replaced with
Johnny> MS-DOS in JARGON.TXT.
Language problem: I was being sarcastic.
Bill Westfield
and can't find a single reference, positive or negative, to either Unix
*or* MS-DOS.
The original IBM-PC was released in AUGUST of 1981.
I don't think the anti-xxxOS wars started in a big way until the old OS's
started dissapearing like dinosaurs. TOPS-10 and TOPS-20 were essentially
the first "operating environments" to lose their support, making the Dec
36bit programmers the first set of grumpy old men.
--
(remove spam food from return address)
Johnny Billquist
>
> >>>>> On Tue, 22 Feb 2000 11:51:56 GMT, Johnny Billquist ("Johnny") writes:
>
> Johnny> Oh? I thought you said that hackers love Unix, and that Unix got replaced with
> Johnny> MS-DOS in JARGON.TXT.
>
> Language problem: I was being sarcastic.
That's what you get when Swedish is your first language.
Too bad. I guess the sarcasm was pretty good then. Wish I had caught
it, though... :-)
jmfb...@aol.com
I wrote a story about it.
/BAH
Subtract a hundred and four for e-mail.
Jim Stewart
jmfb...@aol.com wrote:
>
> >I don't think the anti-xxxOS wars started in a big way until the old OS's
> >started dissapearing like dinosaurs. TOPS-10 and TOPS-20 were essentially
> >the first "operating environments" to lose their support, making the Dec
> >36bit programmers the first set of grumpy old men.
> >
> Actually, they were very, very pissed off not-so-old men.
> I wrote a story about it.
Could you post it?
Jim
Victor Eijkhout
> To quote
> Ken Pier of Xerox PARC:
>
> I liken starting one's computing career with Unix, [...]
> By the time they
> find out differently, it is too late. They already think
> that the writing of shell scripts is a natural act.
What's the alternative to writing shell scripts? I was never hardcore
into it, but I did it in some form on MSDOS, VMS, Unix, Mac (AppleScript),...
The Unix ones have a flavour of their own, but no better or worse than
some other ones.
I'm really puzzled what the above comment means.
--
Victor Eijkhout
"When I was coming up, [..] we knew exactly who the they were. It was us
versus them, and it was clear who the them was were. Today, we are not
so sure who the they are, but we know they're there." [G.W. Bush]
Neil Franklin
Where does one get such an emulator (or even an newer one) from?
Where does one get a ITS disk image from?
Where does one get docs about how to use ITS?
Asked by an Unixer who would like to explore this "so much better"
system to can get a well founded view on what it really is like.
--
Neil Franklin, ne...@franklin.ch.remove http://neil.franklin.ch/
Nerd, Geek, Hacker, Unix Wizzard, Sysadmin, Roleplayer, Mystic
Computer: a toy, speeds work so that you have more time to play
Daniel Seagraves
> Christopher C Stacy <cst...@world.std.com> writes:
>
> > >>>>> On Tue, 22 Feb 2000 20:11:40 GMT, Jay Maynard ("Jay") writes:
> > Jay> I hadn't heard that anyone had ITS running on a -10 emulator...
> >
> > I think that ITS was the first thing to run on the first PDP-10
> > emulator (KLH-10), quite a few years ago.
>
> Where does one get such an emulator (or even an newer one) from?
>
> Where does one get a ITS disk image from?
>
> Where does one get docs about how to use ITS?
Hey, I'm working on it! :P Hold your horses!
"Confuse, annoy, and DEE-STROY!" -- Jet Wolf | "Nothing Happens." -- ADVENT
"...A man can pass his family and his name down through his sons, but it's
his honour that gets passed through his daughters. He can see the best
and worst of life in his girls. A daughter is something far too precious,
and he'll do anything to protect her."
-- Reichsfuehrer Siegfried Koenig, _Matrose_Mond_, David Oliver
Ron Newman
Victor stated...
>
>Tim Shoppa <sho...@trailing-edge.com> writes:
>
>> To quote
>> Ken Pier of Xerox PARC:
>>
>> I liken starting one's computing career with Unix, [...]
>> By the time they
>> find out differently, it is too late. They already think
>> that the writing of shell scripts is a natural act.
>
>What's the alternative to writing shell scripts?
Perl.
--
Ron Newman rne...@thecia.net
http://www2.thecia.net/users/rnewman/home.html
Peter da Silva
>Perl.
How's a different scripting language an alternative to scripting? I mean
perl is a lot more powerful than sh, but it's basically the same sort of
program.
I think the Xerox people have this idea that you shouldn't ever have to
write programs.
--
This is The Reverend Peter da Silva's Boring Sig File - there are no references
to Wolves, Kibo, Discordianism, or The Church of the Subgenius in this document
Executive Vice President, Corporate Communications, Entropy Gradient Reversals.
Tim Shoppa
>
> Tim Shoppa <sho...@trailing-edge.com> writes:
>
> > To quote
> > Ken Pier of Xerox PARC:
> >
> > I liken starting one's computing career with Unix, [...]
> > By the time they
> > find out differently, it is too late. They already think
> > that the writing of shell scripts is a natural act.
>
> into it, but I did it in some form on MSDOS, VMS, Unix, Mac (AppleScript),...
> The Unix ones have a flavour of their own, but no better or worse than
> some other ones.
>
> I'm really puzzled what the above comment means.
Two things:
1. Lots of functions that are "built-in" or "CUSPy" commands
on regular operating systems have to be done with a script under
Unix. Common examples: "RENAME *.DAT *.OLD" (Even MS-DOS gets
that one right), "DELETE/BEFORE=YESTERDAY *.EXE", etc. Anything
that needs the "find" command counts as a script because of all
the lousy punctuation you have to use and the fact that you have
to scroll through all the unreadable man page for "find" before
you get to the examples.
2. Real programmers, of course, just dash off a bit of
macro assembler to do anything complicated. Unix programmers
don't know what a real macro assembler is, and what they think of
as a macro processor ("cpp", C's pre-processor) hardly counts.
Tim.
David Rifkind
>On 23 Feb 2000 11:23:10 -0500, in article <omog9799...@disco.cs.utk.edu>,
>Victor stated...
>>
>>
>>> To quote
>>> Ken Pier of Xerox PARC:
>>>
>>> I liken starting one's computing career with Unix, [...]
>>> By the time they
>>> find out differently, it is too late. They already think
>>> that the writing of shell scripts is a natural act.
>>
>>What's the alternative to writing shell scripts?
>
Talk about unnatural acts....
--
"The privileged being which we call human is distinguished from other
animals only by certain double-edged manifestations which in charity we
can only call 'inhuman.'" -- Epiktistes
Neil Franklin
> On 23 Feb 2000, Neil Franklin wrote:
>
> > Where does one get such an emulator (or even an newer one) from?
> >
> > Where does one get a ITS disk image from?
> >
> > Where does one get docs about how to use ITS?
>
> Hey, I'm working on it! :P Hold your horses!
Great!
Make sure your announce also hits a.f.c when its time arrives.
Peter da Silva
Tim Shoppa <sho...@trailing-edge.com> wrote:
>1. Lots of functions that are "built-in" or "CUSPy" commands
>on regular operating systems have to be done with a script under
>Unix.
And vice-versa. Unless you count "command | grep stuff" and other
super-common idioms as "scripts".
Gerard S.
| Tim Shoppa <sho...@trailing-edge.com> writes:
|
| > To quote
| > Ken Pier of Xerox PARC:
| >
| > I liken starting one's computing career with Unix, [...]
| > By the time they
| > find out differently, it is too late. They already think
| > that the writing of shell scripts is a natural act.
|
| into it, but I did it in some form on MSDOS, VMS, Unix, Mac (AppleScript),...
| The Unix ones have a flavour of their own, but no better or worse than
| some other ones.
|
| I'm really puzzled what the above comment means.
One alternative to shell scripts is REXX --- as a mater of fact, it was
written with this in mind. REXX was, if you will, the shell script for VM/CMS,
and is used very heavily in that operating system.
Gerard S.
Gerard S.
|In article <M3_s4.205$vk....@newsfeed.slurp.net>,
|Gerard S. <ger...@prairietech.net> wrote:
|>One alternative to shell scripts is REXX --- as a mater of fact, it was
|>written with this in mind.
|
|procedural rather than dataflow-oriented like the UNIX shell, and with
|its own set of quirks.
|
|I've used it on the Amiga, and I prefer /bin/sh.
|
|This is The Reverend Peter da Silva's Boring Sig File - there are no references
|to Wolves, Kibo, Discordianism, or The Church of the Subgenius in this document
|
|Executive Vice President, Corporate Communications, Entropy Gradient Reversals.
REXX isn't a different kind of shell, any more than FORTRAN or C would be; it's
a
language. The power of REXX comes in that it can "direct" statements (if you
will) to
the "host", whether it be UNIX or DOS or CMS.
However, as I understand it, some implementations require the use of a shell to
make (or
let) REXX execute correctly. By the way, my understanding is not first-hand.
Gerard S.
Tim Shoppa
>
> In article <M3_s4.205$vk....@newsfeed.slurp.net>,
> Gerard S. <ger...@prairietech.net> wrote:
> >One alternative to shell scripts is REXX --- as a mater of fact, it was
> >written with this in mind.
>
> REXX is simply a different kind of shell, one that is traditional and
> procedural rather than dataflow-oriented like the UNIX shell, and with
> its own set of quirks.
And there's good reason why REXX is procedural and not dataflow-
oriented: REXX was designed to work on machines that manipulate
records and complex data structures. The `pipes' in a Unix shell
fall apart once you leave the tiny subset of things that you
can conveniently do in the "everything is a stream of bytes" model.
Tim.
Ron Newman
da Silva) wrote:
> In article <891ct0$1g...@edrn.newsguy.com>,
> Ron Newman <rne...@thecia.net> wrote:
> >>What's the alternative to writing shell scripts?
>
> >Perl.
>
> How's a different scripting language an alternative to scripting?
It's not. But it's certainly an alternative to *shell* scripts.
Since someone else has brought up REXX, I'd be very interested in
hearing a comparison of Perl and REXX. Is anyone familiar enough
with both to start this discussion?
Marco S Hyman
> (And yes, I write one-off C programs to "do anything complicated".)
Really? I do my one-of-a-kind in awk. Usually goes together much
faster than C. I suppose I could also use perl, but then I'd have
to learn perl and I've always found awk more than adequate.
// marc
Joe Pfeiffer
I think that's why a lot of us do our one-of-a-kind things in C. I
can see in the abstract that awk (which I've written a couple of
scripts in) or Perl (which I haven't, though I've modified a few)
could do the job more easily than C, first I'd have to learn them well
enough to be productive. And by the time I've done that, I've gotten
the C program running.
--
Joseph J. Pfeiffer, Jr., Ph.D. Phone -- (505) 646-1605
Department of Computer Science FAX -- (505) 646-1002
New Mexico State University http://www.cs.nmsu.edu/~pfeiffer
VL 2000 Homepage: http://www.cs.orst.edu/~burnett/vl2000/
David Given
"Paul Grayson" <paul.g...@virgin.net> writes:
[...]
> I stopped reading Slashdot when I realised the mentality of some of the
> posters. I would sometimes post a response to a thread, usually to consider
> a different techology. I'd often receive a response to the effect "Are you
> on crack!", or even worse. I sometimes would receive abusive emails from
> people I'd angered. Sometimes it is not much different to the letters pages
> of early 80s micro magazines. And the majority are ignorant of computer
> history, too!
This is what the moderating system is for. Set your moderation threshold
to 3 and suddenly you'll only see intelligent posts.
...after all, a.f.c gets its share of flamage, too. It's less high-profile
than Slashdot but they're still there. 90% of everone are idiots.
--
+- David Given ---------------McQ-+ "Scuzzy Wuzzy was a bus.
| Work: d...@tao-group.com | Scuzzy Wuzzy caused no fuss.
| Play: dgi...@iname.com | Scuzzy Wuzzy wasn't SCSI, was he?"
+- http://wired.st-and.ac.uk/~dg -+ --- Jordin Kare
Peter da Silva
Gerard S. <ger...@prairietech.net> wrote:
>One alternative to shell scripts is REXX --- as a mater of fact, it was
>written with this in mind.
REXX is simply a different kind of shell, one that is traditional and
procedural rather than dataflow-oriented like the UNIX shell, and with
its own set of quirks.
I've used it on the Amiga, and I prefer /bin/sh.
--
Pete Fenelon
> 2. Real programmers, of course, just dash off a bit of
> macro assembler to do anything complicated. Unix programmers
> don't know what a real macro assembler is, and what they think of
> as a macro processor ("cpp", C's pre-processor) hardly counts.
*REAL* Unix programmers don't need a macro-assembler, because they've
got m4.
pete
--
pe...@fenelon.com "enmimes sont les gougebosquex et le momerade horsgrave!"
Peter Seebach
Tim Shoppa <sho...@trailing-edge.com> wrote:
>1. Lots of functions that are "built-in" or "CUSPy" commands
>on regular operating systems have to be done with a script under
>that one right), "DELETE/BEFORE=YESTERDAY *.EXE", etc. Anything
>that needs the "find" command counts as a script because of all
>the lousy punctuation you have to use and the fact that you have
>to scroll through all the unreadable man page for "find" before
>you get to the examples.
And yet, you *can* do all of these things - many of the things you can
do with simple scripts under Unix can't be done at all in many other
systems.
>2. Real programmers, of course, just dash off a bit of
>macro assembler to do anything complicated. Unix programmers
>don't know what a real macro assembler is, and what they think of
>as a macro processor ("cpp", C's pre-processor) hardly counts.
cpp is *NOT* C's pre-processor. It is a specialized superset of the way some
of the original C implementations provided limited macro functionality. It
is not necessarily correlated closely to any part of a functioning C compiler.
That said, I think I'm a programmer, and I still refuse to touch any kind of
assembler; I don't believe my code should have *any* references to the machine
it's on.
(And yes, I write one-off C programs to "do anything complicated".)
Yeah, it would be nice if you could "easily" do things like 'REN *.DAT *.OLD'
in Unix. It would also be nice if you could "DEL A.TXT B.TXT" in DOS.
Everyone has limits; Unix's strength is that the limits can be worked around.
-s
--
Copyright 2000, All rights reserved. Peter Seebach / se...@plethora.net
C/Unix wizard, Pro-commerce radical, Spam fighter. Boycott Spamazon!
Consulting & Computers: http://www.plethora.net/
Get paid to surf! No spam. http://www.alladvantage.com/go.asp?refid=GZX636
Don Stokes
Tim Shoppa <sho...@trailing-edge.com> wrote:
>records and complex data structures. The `pipes' in a Unix shell
>fall apart once you leave the tiny subset of things that you
>can conveniently do in the "everything is a stream of bytes" model.
Oh, I dunno. Out in VMS-land mailboxes get used a lot as pipes,
even though the OS is agressively record-oriented. Most Unix
applications process input lines rather than characters -- the fact
that they do this by peeling apart a character stream based on LF
delimiters rather than being explicitly handed records doesn't alter
the philosophy behind those applications; they'd drop right into a
record managed environment with very little change.
-- don
Simon Allaway
Peter Seebach wrote:
>
> And yet, you *can* do all of these things - many of the things you can
> do with simple scripts under Unix can't be done at all in many other
> systems.
I've been quite 'pleased' with what can be done with the Windows
Scripting Host. Using that with a simple database has certainly made
administering NT domains a lot easier. And tie that up with a web server
and you've got easily built admin tools.
*shrug* My 2 cents.
Simon
--
--------------------------------
"I'm against animal experiments.
They just get nervous and get
all the answers wrong"
Peter Seebach
Gerard S. <ger...@prairietech.net> wrote:
>REXX isn't a different kind of shell, any more than FORTRAN or C would be; it's
>a
>language.
So is a shell.
Peter da Silva
>oriented: REXX was designed to work on machines that manipulate
>fall apart once you leave the tiny subset of things that you
>can conveniently do in the "everything is a stream of bytes" model.
We could spend a lot of time going back and forth over this, but I'll
just say that I came to UNIX from systems that used a database metaphor
for files and being able to take control of things was like a breath of
fresh air.
It's amazing what you can do with a stream of bytes, particularly once you
catch on to the fact that the UNIX tools *are* database operators...
Peter da Silva
Ron Newman <rne...@thecia.net> wrote:
>> How's a different scripting language an alternative to scripting?
>It's not. But it's certainly an alternative to *shell* scripts.
I've talked with people who've used Perl and Tcl as their shells. Tcl is
in some ways better suited, since tclsh acts like REXX and falls back to
running external programs if it doesn't understand your command... but
without invoking an external argument quoting mechanism.
>Since someone else has brought up REXX, I'd be very interested in
>hearing a comparison of Perl and REXX. Is anyone familiar enough
>with both to start this discussion?
Perl:REXX :: C:Pascal
Matthew W. Miller
>On Tue, 22 Feb 2000 01:28:22 GMT, Christopher C Stacy <cst...@world.std.com>
>wrote:
>>Some people were annoyed by his changing references that denigrated UNIX
>>to refer instead to MS-DOS, since, as we all know, UNIX is what hackers love.
>All of the complaining is from embittered old ITS hackers pissed off that
>their beloved OS has sunk beneath the waves.
No. Personally I'm annoyed by Raymond's nastiness about Commodores, both
in the New Hacker's Dictionary ("notoriously crocky little bitty boxes")
and in the default Linux /etc/termcap ("[Commodore 64's] can still be
found gathering dust in closets everywhere.") I can understand a little
sarcasm, but jeez, you'd think Max Toy spoiled his nice new rattle or
something. Maybe Raymond's one of those weird Atari 800 fans. :)
--
Matthew W. Miller -- ma...@infinet.com
Mark Statzer
> All of the complaining is from embittered old ITS hackers pissed off that
> their beloved OS has sunk beneath the waves.
Amen! Can I get a witness!
> <music href="mp3://Eagles/Get Over It!">
> I miss OS/2, too, but I don't dedicate a chunk of my brainpower to it any
> more...
> </music>
I bite down on my tongue *hard* when I read the ZDNet Talkback sections
pertaining to Microsoft VS DOJ, and the letters that proclaim that
"Microsoft never did anything wrong, you guys are just anti competitive
losers"....
Anyone who worked as hard as I (and others) did cheerfully weeding out
the bugs in OS/2 2.0 and all those betas of 2.1, and can now only fondly
remember the dust that remains ( and see all of those features, stolen
*badly*, in W9X) have a lot of trouble buying the "Bill's innocent, I
tell ya!" line.
It may be true that very little they have done is demonstratively
illegal, but that don't mean it wasn't evil, and harmful to the
industry.
<grumble><bitch>
> ESR's changing of those refs seems to me to make the entries in question
> more universally understandable. Lots of folks like Unix, whether you choose
> to admit and understand it or not. Many, many fewer like MS-DOS.
True, but many more today at least have some *experience* with DOS (not
all traumatic) and don't know bupkus about *nix.
Yes, Virginia, it *is* possible to be a full-bore computer fanatic and
*not* worship @ First Presbyterian *nix...
Mark "remember TeamOS/2?" Statzer
--
*NOTE* I am not responsible for equipment damage due to reeeealy
dumb children with no parental supervision, and access to a hammer.
Johnny Billquist
>
> 2. Real programmers, of course, just dash off a bit of
> macro assembler to do anything complicated. Unix programmers
> don't know what a real macro assembler is, and what they think of
> as a macro processor ("cpp", C's pre-processor) hardly counts.
This brings up one thing that fascinates me.
Why have proper macro handling fallen onto bad times?
Nowadays, everyone seem to think that macro expansion should
be done by a preprocessor.
It's sufficient for C perhaps, but when you write assembler,
it just plainly sucks, yet companies that sell assemblers now
just tuck on a version of cpp or m4.
Have people forgotten how to use a macro assembler, or are
people just cheap?
Johnny
--
Johnny Billquist | johnny.b...@netinsight.net
Net Insight AB | phone: +46 8 685 04 88
Västberga Allé 9 | fax: +46 8 685 04 20
Box 42093 |
SE-126 30 STOCKHOLM, Sweden | http://www.netinsight.net
Johnny Billquist
>
> In alt.folklore.computers Tim Shoppa <sho...@trailing-edge.com> wrote:
> > 2. Real programmers, of course, just dash off a bit of
> > macro assembler to do anything complicated. Unix programmers
> > don't know what a real macro assembler is, and what they think of
> > as a macro processor ("cpp", C's pre-processor) hardly counts.
>
> got m4.
*Ha*! m4 sucks in comparision with a macro assembler. m4 hardly gives
you anything more than cpp does. Both falls miserably short of what
you want.
Mark Crispin
> Anyone who worked as hard as I (and others) did cheerfully weeding out
> the bugs in OS/2 2.0 and all those betas of 2.1, and can now only fondly
> remember the dust that remains ( and see all of those features, stolen
> *badly*, in W9X) have a lot of trouble buying the "Bill's innocent, I
> tell ya!" line.
Imagine that, someone who actually feels sorry for IBM in this newsgroup.
Next thing, we'll be hearing about how wonderful JCL, OS/360, HASP, TSS,
all was until that nasty Microsoft spoiled it with MS-DOS. Let's not
forget AIX, that wonderful variant of UNIX. [venomous sarcasm]
Excuse me while I gag.
I've heard a number of stories of what happened to OS/2. There's a common
thread regarding IBM having shot itself in the foot. IBM fully intended
to break Microsoft, and instead broke itself.
Say what you will about Microsoft being the Evil Empire, but quite
frankly, Microsoft is much more benign than IBM or SUN. Or, for that
matter, DEC. The world is a much better place with the dominant OS not
being in the hands of a hardware vendor.
-- Mark --
* RCW 19.190 notice: This email address is located in Washington State. *
* Unsolicited commercial email may be billed $500 per message. *
Science does not emerge from voting, party politics, or public debate.
Johnny Billquist
>
> On Thu, 24 Feb 2000, Mark Statzer wrote:
> > Anyone who worked as hard as I (and others) did cheerfully weeding out
> > the bugs in OS/2 2.0 and all those betas of 2.1, and can now only fondly
> > remember the dust that remains ( and see all of those features, stolen
> > *badly*, in W9X) have a lot of trouble buying the "Bill's innocent, I
> > tell ya!" line.
>
> Imagine that, someone who actually feels sorry for IBM in this newsgroup.
The evil empore is long forgotten.
> Next thing, we'll be hearing about how wonderful JCL, OS/360, HASP, TSS,
> all was until that nasty Microsoft spoiled it with MS-DOS. Let's not
> forget AIX, that wonderful variant of UNIX. [venomous sarcasm]
>
> Excuse me while I gag.
I think I've already seen people talk fondly over JCL and so on in
a.f.c...
> I've heard a number of stories of what happened to OS/2. There's a common
> thread regarding IBM having shot itself in the foot. IBM fully intended
> to break Microsoft, and instead broke itself.
True.
> Say what you will about Microsoft being the Evil Empire, but quite
> frankly, Microsoft is much more benign than IBM or SUN. Or, for that
> matter, DEC. The world is a much better place with the dominant OS not
> being in the hands of a hardware vendor.
Not sure I agree with you here. Just because all hardware vendors
have proven themself to be a pain in everyones ass don't make MS
a good guy. MS have mistreated everyone just as bad as DEC ever did,
it just turned out that MS isn't as easy to get the blame to stick at.
Ignatios Souvatzis
rne...@thecia.net (Ron Newman) writes:
> Since someone else has brought up REXX, I'd be very interested in
> hearing a comparison of Perl and REXX. Is anyone familiar enough
> with both to start this discussion?
Uh, half the CMS system "programs" I looked at were REXX scripts back then.
And AREXX (Amiga REXX) is certainly powerful enough to write a uucp2smtp
script... that is, a script that searches the uucp spool for rmail jobs
and sends them out through SMTP instead.
Regards,
-is
--
* Progress (n.): The process through which Usenet has evolved from
smart people in front of dumb terminals to dumb people in front of
smart terminals. -- o...@burnout.demon.co.uk (obscurity)
jmfb...@aol.com
<EE0FE242DB77CC31.351302E7...@lp.airnews.net>,
Jim Stewart <jste...@jkmicro.com> wrote:
>
>
>jmfb...@aol.com wrote:
>>
>> >I don't think the anti-xxxOS wars started in a big way until the old
OS's
>> >started dissapearing like dinosaurs. TOPS-10 and TOPS-20 were
essentially
>> >the first "operating environments" to lose their support, making the
Dec
>> >36bit programmers the first set of grumpy old men.
>> >
>> Actually, they were very, very pissed off not-so-old men.
>> I wrote a story about it.
>
>Could you post it?
>
>Jim
I did a while ago. Do the people in the -10 newsgroup mind
if it's posted again? I don't know the netocol about reposting
long stories.
/BAH
Subtract a hundred and four for e-mail.
Daniel Seagraves
> In alt.folklore.computers Tim Shoppa <sho...@trailing-edge.com> wrote:
> > 2. Real programmers, of course, just dash off a bit of
> > macro assembler to do anything complicated. Unix programmers
> > don't know what a real macro assembler is, and what they think of
> > as a macro processor ("cpp", C's pre-processor) hardly counts.
>
> *REAL* Unix programmers don't need a macro-assembler, because they've
> got m4.
No, *REAL* UNIX programmers don't need m4, because they have an
8-bit-clean text editor... ^_^
(Note that I didn't say "hex editor". PDP-11s use octal numbers. :)
"Confuse, annoy, and DEE-STROY!" -- Jet Wolf | "Nothing Happens." -- ADVENT
"...A man can pass his family and his name down through his sons, but it's
his honour that gets passed through his daughters. He can see the best
and worst of life in his girls. A daughter is something far too precious,
and he'll do anything to protect her."
-- Reichsfuehrer Siegfried Koenig, _Matrose_Mond_, David Oliver
Jay Maynard
>No. Personally I'm annoyed by Raymond's nastiness about Commodores, both
>in the New Hacker's Dictionary ("notoriously crocky little bitty boxes")
>and in the default Linux /etc/termcap ("[Commodore 64's] can still be
>found gathering dust in closets everywhere.") I can understand a little
>sarcasm, but jeez, you'd think Max Toy spoiled his nice new rattle or
>something. Maybe Raymond's one of those weird Atari 800 fans. :)
I have a hard time disagreeing with him...between the legacy of
k-kik-k000()o0o()000oool-rad-mcbadical warezzzzzz kiddies with 32-column
Usenet posts, and being gratuitously incompatible with ASCII, the C=64 is a
stain on the history of computing.
Jay Maynard
wrote:
>Imagine that, someone who actually feels sorry for IBM in this newsgroup.
How about feeling sorry for the users who are stuck with Windows, rather
than something that was, in 1994, what Windows 2000 is only now beinning to
reach?
>Next thing, we'll be hearing about how wonderful JCL, OS/360, HASP, TSS,
>all was until that nasty Microsoft spoiled it with MS-DOS. Let's not
>forget AIX, that wonderful variant of UNIX. [venomous sarcasm]
From the mouth of a famous burbler like you, I'd have expected this crap.
No, OS/360 and MVS weren't the great programmer environment of your beloved
ITS. They weren't *meant* to be. They were, and are, bulletproof
environments for doing lots of I/O intensive tasks with supreme
reliability. You know, doing mundane things like printing your paycheck,
along with thousands of others, every couple of weeks. Boring stuff that
doesn't interest hackers, but is just as essential in the grand scheme.
For all of AIX's faults (and I'll agree that they're legion), even they get
a few things right: it's the only variant of Unix I've seen where you can
add a disk and extend the root filesystem on the fly without having to take
the machine down even for a reboot. Nothing is perfect, either good or evil.
>I've heard a number of stories of what happened to OS/2. There's a common
>thread regarding IBM having shot itself in the foot. IBM fully intended
>to break Microsoft, and instead broke itself.
IBM couldn't market ice water to a parched Bedouin camel herder in the
middle of the Sinai at high noon in the middle of August. That doesn't eman
that I can't wish that OS/2 had thrived, and provided a real alternative to
M$. I run Windows 2000 now not because I want to, but because I have to.
>Say what you will about Microsoft being the Evil Empire, but quite
>frankly, Microsoft is much more benign than IBM or SUN. Or, for that
>matter, DEC. The world is a much better place with the dominant OS not
>being in the hands of a hardware vendor.
Right. Uh huh.
I don't suppose it's occurred to you that the dominant OS *can't* be in the
hands of a hardware vendor any more? A Windows *couldn't* achieve the
monopoly status it has if it were owned by Compaq, or Dell, or IBM...
David Given
Johnny Billquist <Johnny.B...@netinsight.se> writes:
> Tim Shoppa wrote:
>> 2. Real programmers, of course, just dash off a bit of
>> macro assembler to do anything complicated. Unix programmers
>> don't know what a real macro assembler is, and what they think of
>> as a macro processor ("cpp", C's pre-processor) hardly counts.
>
> Why have proper macro handling fallen onto bad times?
>
> Nowadays, everyone seem to think that macro expansion should
> be done by a preprocessor.
> It's sufficient for C perhaps, but when you write assembler,
> it just plainly sucks, yet companies that sell assemblers now
> just tuck on a version of cpp or m4.
> Have people forgotten how to use a macro assembler, or are
> people just cheap?
Well, I agree completely. We have a macro assembler that we use in-house;
it's damn good, too, even if the syntax did tend to have grown rather than
been designed. Before I came here I'd never met a decent macro assembler.
As Tim said, I thought cpp was one.
Now I've met it, and used it, and written some truly horrific macros in
it, I can't believe that I had ever managed to survive without one. It's
just so *useful*. You wouldn't believe it, but the ability to write
turing-complete programs in your assembler metalanguage is something you
use more than you'd expect.
Are there any decent macro assembler preprocessors for Unix? Could I get
away with plugging one into the front of my C compiler?
--
+- David Given ---------------McQ-+
| Work: d...@tao-group.com | All things considered, insanity may be the
| Play: dgi...@iname.com | only reasonable alternative.
+- http://wired.st-and.ac.uk/~dg -+
Brian Inglis
wrote:
>In article <891ct0$1g...@edrn.newsguy.com>,
>Ron Newman <rne...@thecia.net> wrote:
>>>What's the alternative to writing shell scripts?
>
>>Perl.
>
>How's a different scripting language an alternative to scripting? I mean
>perl is a lot more powerful than sh, but it's basically the same sort of
>program.
>
>I think the Xerox people have this idea that you shouldn't ever have to
>write programs.
Yeah -- just do everything manually on the computer GUI, just
like you would if you didn't have a computer, but slower! What a
waste! Why bother! Where's the leverage! Computers are for
automation of repetitive or difficult (for humans) tasks! Notice
that any wi[l]d[e]ly successful product has some kind of
programming language in/on/around it! That's been the one thing
holding Windows back from real popularity -- no scripting! Users
are limited to banging on the keys in a single application or
using a macro language in one application! Imagine where we might
be if MS had concentrated on a user scripting language instead of
GUI bells and whistles! Whew -- narrow escape!
I REALLY wanted to use ALL CAPS here, but kept a lid on it with a
bunch of bangs!
Thanks. Take care, Brian Inglis Calgary, Alberta, Canada
--
Brian_...@CSi.com (Brian dot Inglis at SystematicSw dot ab dot ca)
use address above to reply
Brian Inglis
<sho...@trailing-edge.com> wrote:
>Victor Eijkhout wrote:
>>
>> Tim Shoppa <sho...@trailing-edge.com> writes:
>>
>> > To quote
>> > Ken Pier of Xerox PARC:
>> >
>> > I liken starting one's computing career with Unix, [...]
>> > By the time they
>> > find out differently, it is too late. They already think
>> > that the writing of shell scripts is a natural act.
>>
>> What's the alternative to writing shell scripts? I was never hardcore
>> into it, but I did it in some form on MSDOS, VMS, Unix, Mac (AppleScript),...
>> The Unix ones have a flavour of their own, but no better or worse than
>> some other ones.
>>
>> I'm really puzzled what the above comment means.
>
>Two things:
>
>1. Lots of functions that are "built-in" or "CUSPy" commands
>on regular operating systems have to be done with a script under
>Unix. Common examples: "RENAME *.DAT *.OLD" (Even MS-DOS gets
Lots of functions that are built in commands or regular operating
systems have to be done by manually selecting and clicking and
button pushing on GUIs.
Anything you can do in a shell script, you can do on a shell
command line! A good command line editor shell helps, but
anything better than a bare Bourne's shell works. Of course a lot
of things that you can do on the UNIX command line, you can't do
easily, if at all, on other systems, without writing scripts or
programs! Your example is a one liner, that I normally
parameterize in a shell command alias:
ls *.dat | sed 's/\([^\.]*\).dat/mv \1.dat \1.old/' | sh -s
>that one right), "DELETE/BEFORE=YESTERDAY *.EXE", etc. Anything
Also:
find . -name '*.exe' -mtime +1 -print | xargs rm
or
find . -name '*.exe' -mtime +1 -exec rm {} \; [slow]
>that needs the "find" command counts as a script because of all
>the lousy punctuation you have to use and the fact that you have
Having learned a few useful find options, just as obvious to me
as /BEFORE, and with a delta date format I *can* remember, I can
then do anything on files accessed, modified or changed before or
after yesterday, or any other period, including applying scripts,
and not be limited to just runnning the few predefined commands
that support that option. And I can do it on any OS (that has the
underlying functionality to support my tool set) -- this means
yours!
>to scroll through all the unreadable man page for "find" before
>you get to the examples.
As opposed to navigating the "interactive" help --
help
Specifications
Dates
Delta Dates
Format
-- learn a few pager (pg/more) search commands (e.g. /modified)
to avoid scrolling.
Isn't "prog1 | prog2" a much nicer and more productive notation,
without all the extraneous verbiage:
$ ASSIGN/USER_MODE temp.tmp SYS$OUTPUT
$ RUN PROG1
$ DEASSIGN/USER_MODE SYS$OUTPUT
$ ASSIGN/USER_MODE temp.tmp SYS$INPUT
$ RUN PROG2
$ DEASSIGN/USER_MODE SYS$INPUT
>2. Real programmers, of course, just dash off a bit of
>macro assembler to do anything complicated. Unix programmers
>don't know what a real macro assembler is, and what they think of
>as a macro processor ("cpp", C's pre-processor) hardly counts.
Only when the "real programmer" ends up stuck on the same old,
slow (when bought) piece of hardware and OS for years. Real
programmers can program on any platform, environment, OS, in
whatever language they happen to have at hand, that day, even if
they never saw the language or hardware before. And use all
available processors/systems, if they need them to do the job.
Even today, given a chance and the right tools, you can do a lot
in 1-4K (8 bit) bytes, excluding RTLs, OS libraries, kernel
calls, and the other overhead that code is laid prone by
nowadays.
Unix programmers just dash off a few calls to library routines to
do anything; know that an assembler is where all your source ends
up for translation into object code, whatever language it may be
written in or platform it's written on; that cpp is the C macro
preprocessor; that m4 is the macro processor, and has nothing to
do with assembly. Macro processors bear the same relationship to
macro assemblers as do high level languages to assemblers. But
real macro processors are often more cryptic than C -- about on a
par with TECO. [Gleeful chuckle]
ObAFC: see thread about the TRAC macro processor for a taste.
>Tim.
(Ref card carrying member of the 11/15/32/86/370 fraternities)
Brian Inglis
<sho...@trailing-edge.com> wrote:
>Peter da Silva wrote:
>>
>> In article <M3_s4.205$vk....@newsfeed.slurp.net>,
>> Gerard S. <ger...@prairietech.net> wrote:
>> >One alternative to shell scripts is REXX --- as a mater of fact, it was
>> >written with this in mind.
>>
>> REXX is simply a different kind of shell, one that is traditional and
>> procedural rather than dataflow-oriented like the UNIX shell, and with
>> its own set of quirks.
>
>And there's good reason why REXX is procedural and not dataflow-
>oriented: REXX was designed to work on machines that manipulate
>records and complex data structures. The `pipes' in a Unix shell
>fall apart once you leave the tiny subset of things that you
>can conveniently do in the "everything is a stream of bytes" model.
REXX is a language designed to make common programming types of
jobs (including shell type scripts) easier to learn and to do,
according to Mike C. REXX works on lines of data, that may be
implemented as records on some platforms, and communicates with
the host OS to make the programmer's job easier. REXX may be
regarded as a shell scripting language, but is not a shell in the
sense of UNIX or DCL, and requires suitable shell or OS
interfaces to work.
Pipes are process communication and data flow mechanisms, like
VMS mailboxes or temporary files (which are sometimes used to
implement them), that require a standardized, modular approach to
programming, which is always a plus.
The stream of bytes model is a file storage architecture that
leaves it up to the application to interpret the data.
Application level libraries may do what they like to access the
data, including direct access if supported by the device, and
bypassing the FS and OS layers if possible and desired for
performance, with no changes in the application I/O code or
library code (application data management code may require a
tweak).
The days of flakey, unreliable application data file I/O systems
with write only characteristics are over. Structured data is
stored in databases, unstructured data is stored in text files.
Text files are used as the universal communication and exchange
mechanism between databases. Binary data is used internally and
only on the same platform. Dump the data from your VAX/VMS system
today, load it into your Alpha/OpenVMS or Sun/Solaris or ...
system later today. Change your server host name and you're back
in business.
Binary, record oriented files just limit your ability to upgrade
to the bigger, better, faster, cheaper, more reliable hardware
you need to scale up drastically to compete next year. Do we know
or care what hardware or OS our databases run on nowadays -- only
if they run proprietary, non-portable, non-scalable software!
>Tim.
(Ref card carrying member of BAT/DCL/[bck]sh/REXX fraternities)
Ben Harris
Victor Eijkhout <eijk...@disco.cs.utk.edu> wrote:
>Tim Shoppa <sho...@trailing-edge.com> writes:
>
>> To quote
>> Ken Pier of Xerox PARC:
>>
>> I liken starting one's computing career with Unix, [...]
>> By the time they
>> find out differently, it is too late. They already think
>> that the writing of shell scripts is a natural act.
>
>What's the alternative to writing shell scripts? I was never hardcore
>into it, but I did it in some form on MSDOS, VMS, Unix, Mac (AppleScript),...
>The Unix ones have a flavour of their own, but no better or worse than
>some other ones.
>
>I'm really puzzled what the above comment means.
I'd assume the comment was from someone used to dealing with a LISP system
or somesuch, where the system programming language is also usable for
writing quick scripts in, so the transition from quick script to real
program is a lot easier.
--
Ben Harris
Unix Support, University of Cambridge Computing Service.
If I wanted to speak for the University, I'd be in ucam.comp-serv.announce.
Peter da Silva
Incorporated (though I still prefer the FIG implementation).
Peter da Silva
>I did a while ago. Do the people in the -10 newsgroup mind
>if it's posted again? I don't know the netocol about reposting
>long stories.
GA -o-
Penny Century
>Tim Shoppa <sho...@trailing-edge.com> writes:
>
>> To quote
>> Ken Pier of Xerox PARC:
>>
>> I liken starting one's computing career with Unix, [...]
>> By the time they
>> find out differently, it is too late. They already think
>> that the writing of shell scripts is a natural act.
>
>What's the alternative to writing shell scripts? I was never hardcore
>into it, but I did it in some form on MSDOS, VMS, Unix, Mac (AppleScript),...
This complaint usually seems to come from Lisp Machine users, although
Xerox Smalltalk users and others are equally justified. The idea is that
you have the same (good) language for communicating with all levels of
the system.
The closest widely used approximation to this philosophy these days is
probably Emacs: I don't write Emacs "scripts", I _program_ Emacs and
I have all of its facilities and data-structures at my disposal when I
do so.
If you think of the UNIX shell as a programming language whose operators
are (mostly) programs it doesn't (or shouldn't) take long to wonder why
UNIX programs almost exclusively communicate by
a) returning integers
b) writing strings to their output.
Many people think that a richer set of data-types is desirable for optimal
programming convenience.
In fact you could make a case for the current proliferation of "scripting"
languages and "distributed component architectures" as a belated (and largely
half-baked) admission by the mainstream of computing that the Lisp and
Smalltalk communities were right all along.
Des,
who still thinks Emacs is a superior environment to COM and Visual Basic.
Pete Fenelon
> Uh, half the CMS system "programs" I looked at were REXX scripts back then.
Biggest and nastiest thing I ever saw written in REXX was a complete IRC
client for VM/CMS with TCP/IP installed. It looked like it hooked into
some of the screen-handling stuff (ISPF?) on forms-mode terminals. *Deep*
scariness.
pete
--
pe...@fenelon.com "enmimes sont les gougebosquex et le momerade horsgrave!"
Pete Fenelon
> It's amazing what you can do with a stream of bytes, particularly once you
> catch on to the fact that the UNIX tools *are* database operators...
I've seen at least two RDBMSs written entirely out of shell scripts and the
tools you'd find if you took a standard UNIX machine out of the box and
installed nothing optional. Slow (one of them took about ten seconds to
insert a record into a table on a VAX 11/750!), but definitely an RDBMS, and
definitely built from simple tools... so had the great advantage that *you
could read all the files yourself*! There were no "green bytes", all the
"meta-data" was stored in the same format as the data itself... marvellous!
Death to obscure binary formats!
The name tdb seems to come to mind, and I vaguely associate it with
Prof Harold Thimbleby... I know there was another one later.
Pete Fenelon
> se...@plethora.net (Peter Seebach) writes:
>> (And yes, I write one-off C programs to "do anything complicated".)
> Really? I do my one-of-a-kind in awk. Usually goes together much
> faster than C. I suppose I could also use perl, but then I'd have
> to learn perl and I've always found awk more than adequate.
I've never found a reason for using Perl -- apart from needing it around
to install and/or configure the mounds of halfwit Unix-like freeware around
there these days. Most of the simple stuff Perl can do I find much neater
in awk/sed/sh and most of the difficult stuff I'd rather write in a
robust compiled language that's near-standard, rather than some
ad-hoc language controlled by one man. And, when it comes down to it,
I find the Perl enthusiast mindset ("let's see how many language features
I can use! Hey, I may not know APL but I can be obscure!") tiresome.
I write one-offs in either C, C++ or a combination of sh, awk, sed, tr,
etc. etc., and possibly other bits of C and C++ depending on a number
of variables --
- the size of the job
- the likelihood that the "one off" will remain "one off"
- the size of the data set I need to work with
- the time available to complete the job
Peter Seebach
Penny Century <pe...@nutri-matic.demon.co.uk> wrote:
>Many people think that a richer set of data-types is desirable for optimal
>programming convenience.
It has pros and cons.
The more data types you may encounter, the harder it is to be sure what you
mean. :)
I like the fact that I can write a program-for-using-other-programs and
it can work with pretty much any Unix utility without changes. Protocol
can be worth it.