No Algol for the STRETCH?

[Quadibloc]

On my web site, at

http://www.quadibloc.com/comp/cp02.htm

on the bottom of the page, I finally illustrate, side by side, the
character set for the IBM 7030 computer, also known as the STRETCH,
and the popular if non-standard Armenian-language character set
ARMSCII. Both of these had the unique characteristic that lower-case
and upper-case characters were interleaved - in hopes of improving
collation.

Anyways, because the STRETCH character set had so many of the
characters needed for ALGOL, I put that diagram on this page rather
than some other page.

Which drives me to ask - Bitsavers only has a FORTRAN manual for the
STRETCH - did IBM or even a customer ever bother to write an ALGOL
compiler for that computer, given the contents of its 120 character
print train?

Or was there only COLASL - which was an ALGOL variant - to make use of
those characters?

John Savard

[Quadibloc]

Couldn't find one by looking here:
http://www.softwarepreservation.org/projects/ALGOL/algol60impl/

John Savard

[Charlie Gibbs]

In article
<5d39d9d6-9099-4098...@yb1g2000pbc.googlegroups.com>,

jsa...@ecn.ab.ca (Quadibloc) writes:

> On my web site, at
>
> http://www.quadibloc.com/comp/cp02.htm
>
> on the bottom of the page, I finally illustrate, side by side, the
> character set for the IBM 7030 computer, also known as the STRETCH,
> and the popular if non-standard Armenian-language character set
> ARMSCII. Both of these had the unique characteristic that lower-case
> and upper-case characters were interleaved - in hopes of improving
> collation.

I've always wondered whether it would be better to do it that way -
but I guess the world was so bound to all upper case at the time
that it was pretty much inconceivable.

Lt. Hauk: You have a problem with that, young man?
Garlick: Absolutely not. I live to collate, sir.
Lt. Hauk: Good. So do I.
-- Good Morning, Vietnam

--
/~\ cgi...@kltpzyxm.invalid (Charlie Gibbs)
\ / I'm really at ac.dekanfrus if you read it the right way.
X Top-posted messages will probably be ignored. See RFC1855.
/ \ HTML will DEFINITELY be ignored. Join the ASCII ribbon campaign!

[Quadibloc]

On Apr 30, 1:00 am, "Charlie Gibbs" <cgi...@kltpzyxm.invalid> wrote:

> I've always wondered whether it would be better to do it that way -
> but I guess the world was so bound to all upper case at the time
> that it was pretty much inconceivable.

Bob Bemer, at least, felt the decision to do this for the STRETCH was
a terrible mistake. And I have to agree; basically, changing an
earlier letter from small to capital should not outweigh changing a
later letter to a different letter. Upper to lower case changes have
to be minor to the character values in the entire string. So this
interleaved case sequence only gives the illusion of doing something
useful.

John Savard

[Patrick Scheible]

On the other hand for a quick and dirty sort based on lexical order
(which is all we get from most software even now in directory listings,
etc.) that would have been a great improvement over having the uppercase
alphabets sorting completely separately from the lowercase alphabets.
And it still wouldn't have been any greater obstacle to a smarter sort
routine if one were wanted.

-- Patrick

[Charles Richmond]

"Patrick Scheible" <k...@zipcon.net> wrote in message
news:86r4hrd...@chai.my.domain...

By "upercase alphabets sorting completely separately from the lowercase
alphabets"... do you mean case-insensative sorting???

--

numerist at aquaporin4 dot com

[Quadibloc]

On Apr 30, 3:58 pm, "Charles Richmond" <numer...@aquaporin4.com>
wrote:

> By "upercase alphabets sorting completely separately from the lowercase
> alphabets"... do you mean case-insensative sorting???

No, he meant sorting using normal string sorting applied to mixed-case
strings. So if one had upper case strings, they would sort from A-Z,
completely separate from the lower-case strings sorting from a-z after
them. I would have thought that obvious from context.

John Savard

[Patrick Scheible]

No, the opposite. I mean if you sort a bunch of words, some of
which start with lowercase and others start with uppercase, all the
uppercase ones sort first and all the lowercase ones sort after all the
uppercase ones. In my opinion that's suboptimal and it would be better
to do case insensitive sorting or distinguishing case only as a less
significant element in the sort than the letter.

-- Patrick

[hanc...@bbs.cpcn.com]

On Apr 30, 5:00 pm, Patrick Scheible <k...@zipcon.net> wrote:

> On the other hand for a quick and dirty sort based on lexical order
> (which is all we get from most software even now in directory listings,
> etc.) that would have been a great improvement over having the uppercase
> alphabets sorting completely separately from the lowercase alphabets.
> And it still wouldn't have been any greater obstacle to a smarter sort
> routine if one were wanted.

That was still in the punched card era, and I suspect the Stretch and
S/360 character sets and architecture were based on punched card
thinking and codes (eg "Move Zone" commands).

In those days, there was so little upper/lower case machine work that
I think they added lower case for future use, not really thinking too
hard about applications (like the sort collating sequence for mixed
case).. Indeed, I think for Stretch particularly they were focused on
heavy duty number crunching, not so much text processing in u/l case.

[hanc...@bbs.cpcn.com]

On Apr 30, 7:05 pm, Patrick Scheible <k...@zipcon.net> wrote:

> No, the opposite.  I mean if you sort a bunch of words, some of
> which start with lowercase and others start with uppercase, all the
> uppercase ones sort first and all the lowercase ones sort after all the
> uppercase ones.  In my opinion that's suboptimal and it would be better
> to do case insensitive sorting or distinguishing case only as a less
> significant element in the sort than the letter.

When I did sorting in Quick Basic I converted the sort field to all
upper case with the UCASE$ function for sorting purposes. In this way
everything came out the same.

In the old days of mainframes everything was converted to upper case
(and often still is), so it wasn't an issue. I'm not sure how they
handle alphabetized lists today with surnames like McBrian or van
Wyck.

[Andrew Swallow]

I would have expected Stretch to have used the ITA2 character code used
for telex and telegrams.

Andrew Swallow

[Patrick Scheible]

In my experience, pretty poorly, with lots of duplicate records.
Library catalogers make a good attempt, with cross-references. Typical
databases -- banks, mailing lists, etc. -- have messy records and try to
use something else as the search key.

-- Patrick

[Quadibloc]

On Apr 30, 8:40 pm, Andrew Swallow <am.swal...@btinternet.com> wrote:

> I would have expected Stretch to have used the ITA2 character code used
> for telex and telegrams.

The PDP-1 and PDP-5 did that, but with BCDIC IBM already had a better
code. So that's the last thing I would have expected; IBM was
competent to make its own peripherals, and didn't have to go aroumd
buying Model 19s or Flexowriters.

John Savard

[hanc...@bbs.cpcn.com]

On Apr 30, 10:40 pm, Andrew Swallow <am.swal...@btinternet.com> wrote:

> I would have expected Stretch to have used the ITA2 character code used
> for telex and telegrams.

That's only 5 bit and requires a shift between letters and numerals.

[Andrew Swallow]

Use 6 bits and you can store the data without the shift.

The usefulness of ITA2 depends on how many telegrams you expect to
process, to the nearest million.

Andrew Swallow

[Quadibloc]

On May 1, 1:06 pm, Andrew Swallow <am.swal...@btinternet.com> wrote:

> The usefulness of ITA2 depends on how many telegrams you expect to
> process, to the nearest million.

Well, yes. And users of the 7030 didn't process telegrams with it.
They used it to do things like simulate the behavior of atomic bombs.

Now, there was the highly-classified 7950, which consisted of a
STRETCH and a character coprocessor. Maybe the software for that
machine did avoid translating 5-level code telegrams into the normal
STRETCH code, but instead processed them natively. But there was no
reason to scream about the purpose of that machine by modifying the
character code used by normal 7030s to match whatever it was doing.

John Savard

[Quadibloc]

On Apr 30, 7:45 pm, hanco...@bbs.cpcn.com wrote:

> When I did sorting in Quick Basic I converted the sort field to all
> upper case with the UCASE$ function for sorting purposes.  In this way
> everything came out the same.
>
> In the old days of mainframes everything was converted to upper case
> (and often still is), so it wasn't an issue. I'm not sure how they
> handle alphabetized lists today with surnames like McBrian or van
> Wyck.

The "proper" way to collate strings would be:

a) pad them all to the same length;
b) concatenate the ucase$ version with the raw version

...that will sort correctly.

John Savard

[Charles Richmond]

"Quadibloc" <jsa...@ecn.ab.ca> wrote in message
news:1e5bb597-2e7b-4f46...@pl9g2000pbb.googlegroups.com...

Perhaps obvious to most... to me, this is unncessary to point out. This is
how what I would consider "normal" sorting to be done with ASCII or
EBCDIC... I thought he must be trying to point out something special.

[Charles Richmond]

"Patrick Scheible" <k...@zipcon.net> wrote in message

news:86a9ofv...@chai.my.domain...

Agreed. And there are facilities that do this sort of thing. The Mac will
sort files in a folder this way.

[Robin Vowels]

On May 1, 2:37 am, Quadibloc <jsav...@ecn.ab.ca> wrote:
> On Apr 30, 1:00 am, "Charlie Gibbs" <cgi...@kltpzyxm.invalid> wrote:
>
> > I've always wondered whether it would be better to do it that way -
> > but I guess the world was so bound to all upper case at the time
> > that it was pretty much inconceivable.
>
> Bob Bemer, at least, felt the decision to do this for the STRETCH was
> a terrible mistake. And I have to agree; basically, changing an
> earlier letter from small to capital should not outweigh changing a
> later letter to a different letter. Upper to lower case changes have
> to be minor to the character values in the entire string.

It's trivial to change from upper to lower and vice versa
by using the TR instruction.
Changing any character from upper to lower and vice versa
would involve adding/subtracting 1 to/from the character value.

The ability to collate properly wold have been a distinct advantage.

[Robin Vowels]

On May 2, 5:06 am, Andrew Swallow <am.swal...@btinternet.com> wrote:

> On 01/05/2013 17:39, hanco...@bbs.cpcn.com wrote:> On Apr 30, 10:40 pm, Andrew Swallow <am.swal...@btinternet.com> wrote:
>
> >> I would have expected Stretch to have used the ITA2 character code used
> >> for telex and telegrams.
>
> > That's only 5 bit and requires a shift between letters and numerals.
>
> Use 6 bits and you can store the data without the shift.

That only gives you 64 characters, and no lower-case.

[Robin Vowels]

On May 2, 5:32 am, Quadibloc <jsav...@ecn.ab.ca> wrote:

> The "proper" way to collate strings would be:
>
> a) pad them all to the same length;

When comparing two strings, it's only necessary to pad the shorter
string
with one blank character.

> b) concatenate the ucase$ version with the raw version

why?

[hanc...@bbs.cpcn.com]

On May 1, 3:06 pm, Andrew Swallow <am.swal...@btinternet.com> wrote:

I don't Stretch was ever used for communications processing.

Western Union did eventually utlize computers for store and forward
processing and routing, but I don't think they used IBM equipment.
(Think they used Univacs).

In the Western Union Technical Review article on ASCII, they weren't
very impressed with it, preferring Baudot. IMHO this came to hurt
them later on--despite all their research into broadband, my
impression is that their network in the 1970s remained mostly low
speed Baudot despite using ASCII for some major systems. One
certainly could use such low speed Baudot liens for data communication
between computers, and it was cheap to do so, but awfully slow
compared to Bell ordinary voice grade lines, let alone conditioned
private lines.
http://massis.lcs.mit.edu/telecom-archives/archives/technical/western-union-tech-review/18-2/p050.htm

[Quadibloc]

So that when two strings differ only in capitalization, a consistent
rule will be used to sort those cases. If De La Rue comes before De la
Rue, then that must be because it compares lower, so that it will also
be true that Dan Van Oosterhuis will come before, and not after, Dan
van Oosterhuis.

John Savard

[Lawrence Statton]

ITA2 only had fifty-some distinct glyphs and no lower-case.

[Andrew Swallow]

Telegrams were all upper case.

Andrew Swallow

[Ahem A Rivet's Shot]

For selected values of correctly. Sorting strings correctly is a
complex business in properly internationalised software - for example
French alphabetic sorting requires comparing from the back of the strings if
the only difference is in the accents. In some languages there are
different conventions for dictionary ordering compared to name ordering.
The ICU library has the most complete implementation of internationalised
sorting rules I know of.

--
Steve O'Hara-Smith | Directable Mirror Arrays
C:>WIN | A better way to focus the sun
The computer obeys and wins. | licences available see
You lose and Bill collects. | http://www.sohara.org/

[Dan Espen]

Ahem A Rivet's Shot <ste...@eircom.net> writes:

> For selected values of correctly. Sorting strings correctly is a
> complex business in properly internationalised software - for example
> French alphabetic sorting requires comparing from the back of the strings if
> the only difference is in the accents.

Say it isn't so.

According to this page:

http://www.unicode.org/reports/tr10/

that applies to Canadian French.

Seems to me, the correct way to deal with that rule is to refuse
to implement it on any computer system. It's just wrong.
Even for people it's wrong.

--
Dan Espen

[Morten Reistad]

In article <2Z6dnSicMLTeUR_M...@bt.com>,

There was/is at least three shift sets in IA2, letters, lowercase and
figures shift. Lowercase is an addon from ca 1965 though.

-- mrr

[Jaimie Vandenbergh]

On Thu, 02 May 2013 18:12:19 -0400, Dan Espen <des...@verizon.net>
wrote:

Said like a true CS engineer. To which as a software project lead I'd
reply "Don't be so provincial. It'll come back as a bug and then
you'll need to implement it anyway."

If you're sorting for use by people, you need to sort the way the
people expect. Or get them to agree and sign off on the spec
describing the way you're sorting it.

Cheers - Jaimie
--
Okay, it works now. Or at least it malfunctions in all the expected ways.
-- Mark Edwards, asr

[hanc...@bbs.cpcn.com]

On May 2, 4:54 pm, Morten Reistad <fi...@last.name> wrote:

> >Telegrams were all upper case.
>
> There was/is at least three shift sets in IA2, letters, lowercase and
> figures shift. Lowercase is an addon from ca 1965 though.

How did a teleprinter handle three levels of shifting? Dot matrix
printing? In 1965, most teleprinters in service were older models,
still using typebars like a tyepwriter.

Was there an additional control character for third shift?

Was this character set widely used in the US?

I don't think "modern" TTY's--the Baudot devices used by the deaf--
make use of a three level code.

[hanc...@bbs.cpcn.com]

On May 2, 4:27 pm, Andrew Swallow <am.swal...@btinternet.com> wrote:

> Telegrams were all upper case.

In the fine 1950s movie, "Executive Suite", the CEO sends a telegram
and it costs him 85c. If he had telephoned his message it would've
cost even more. That would be about $8 in today's money.

Today of course he could send a text, even if charged a la carte, it'd
be only 25c. He could probably send a free fax from his hosts, or use
his cellphone.

Also in the movie they need to get a message to someone on a trip.
They telephone a nearby turnpike interchange and ask them to stop the
man as he comes by. Can you imagine asking a toll facility to do that
today?

E/S was a good movie, and probably could be remade today and easily
brought up to date, including reflecting modern technology in
communications and manufacturing. The attempted stock manipulations
could remain exactly the same.

I still think a movie about the Watsons would make a good drama, plus
offer the chance to have neat shots of punched cards flying through
their machines. (Too bad Mitch Miller isn't around to lead a men's
chorus in Ever Onward).

[Gene Wirchenko]

On Thu, 2 May 2013 07:53:06 -0700 (PDT), Robin Vowels
<robin....@gmail.com> wrote:

>On May 2, 5:32 am, Quadibloc <jsav...@ecn.ab.ca> wrote:
>
>> The "proper" way to collate strings would be:
>>
>> a) pad them all to the same length;
>
>When comparing two strings, it's only necessary to pad the shorter
>string
>with one blank character.

Why? "abc" and "abc " are not equal. If you reach the end of a
string, then stop. The shorter string comes first.

[snip]

Sincerely,

Gene Wirchenko

[Peter Flass]

"It all depends" (tm) In many languages other than the nameless
one-letter one. the shorter string is padded with blanks, or treated as
being padded, to the length of the longer one. This normally doesn't
matter, since no printable characters precede space in any character set
(AFAIK), but it might matter when comparing arbitrary strings of bytes.


--
Pete

[Peter Flass]

On 5/2/2013 10:34 PM, Jaimie Vandenbergh wrote:
> On Thu, 02 May 2013 18:12:19 -0400, Dan Espen <des...@verizon.net>
> wrote:
>
>> Ahem A Rivet's Shot <ste...@eircom.net> writes:
>>
>>> For selected values of correctly. Sorting strings correctly is a
>>> complex business in properly internationalised software - for example
>>> French alphabetic sorting requires comparing from the back of the strings if
>>> the only difference is in the accents.
>>
>> Say it isn't so.
>>
>> According to this page:
>>
>> http://www.unicode.org/reports/tr10/
>>
>> that applies to Canadian French.
>>
>> Seems to me, the correct way to deal with that rule is to refuse
>> to implement it on any computer system. It's just wrong.
>> Even for people it's wrong.
>
> Said like a true CS engineer. To which as a software project lead I'd
> reply "Don't be so provincial. It'll come back as a bug and then
> you'll need to implement it anyway."
>
> If you're sorting for use by people, you need to sort the way the
> people expect. Or get them to agree and sign off on the spec
> describing the way you're sorting it.
>

Unfortunately you can no longer get away with telling them "because
that's the way computers work, we have to do it that way." Used to be a
very effective response.


--
Pete

[Jaimie Vandenbergh]

It was a lousy response then too.

Cheers - Jaimie
--
Moral indignation is jealousy with a halo
- H G Wells

[hanc...@bbs.cpcn.com]

On May 3, 7:56 am, Peter Flass <Peter_Fl...@Yahoo.com> wrote:

> Unfortunately you can no longer get away with telling them "because
> that's the way computers work, we have to do it that way."  Used to be a
> very effective response.

Huh?

In the old days, management and staff could be very resistant when
replacing a manual process with a computer and having it function
worse than the manual system did. Certainly this did happen, but
there were recriminations; new systems could be scrapped as
unworkable. As experience was gained by both computer people and end
users, system quality grew. Remember than in the old days computer
folks were often employees of the same company as the end users so
there was more accountability, including after implementation.

Today, we depend heavily on canned software and fleeting consultants,
so we're more bound to "this is how they do it".

(Obviously there are exceptions).

[Patrick Scheible]

Jaimie Vandenbergh <jai...@sometimes.sessile.org> writes:

> On Thu, 02 May 2013 18:12:19 -0400, Dan Espen <des...@verizon.net>
> wrote:
>
>>Ahem A Rivet's Shot <ste...@eircom.net> writes:
>>
>>> For selected values of correctly. Sorting strings correctly is a
>>> complex business in properly internationalised software - for example
>>> French alphabetic sorting requires comparing from the back of the strings if
>>> the only difference is in the accents.
>>
>>Say it isn't so.
>>
>>According to this page:
>>
>>http://www.unicode.org/reports/tr10/
>>
>>that applies to Canadian French.
>>
>>Seems to me, the correct way to deal with that rule is to refuse
>>to implement it on any computer system. It's just wrong.
>>Even for people it's wrong.
>
> Said like a true CS engineer. To which as a software project lead I'd
> reply "Don't be so provincial. It'll come back as a bug and then
> you'll need to implement it anyway."
>
> If you're sorting for use by people, you need to sort the way the
> people expect. Or get them to agree and sign off on the spec
> describing the way you're sorting it.

Exactly. It's not the way I would sort it, and it's not the even the
way the English-speaking library community sorts things. but then
English doesn't make much use of diacritical marks. I have heard before
that the French library community DOES sort by diacritical marks, as a
less significant sorting element than the letters. Think of the
customer's needs first.

-- Patrick

[Patrick Scheible]

And made a whole lot of people hate computers.

-- Patrick

[Patrick Scheible]

hanc...@bbs.cpcn.com writes:

> Also in the movie they need to get a message to someone on a trip.
> They telephone a nearby turnpike interchange and ask them to stop the
> man as he comes by. Can you imagine asking a toll facility to do that
> today?

I can't believe that happening in real life even in the 1950s. Maybe if
it was law enforcement that asked, or a substantial bribe changed hands.

-- Patrick

[Quadibloc]

On May 3, 10:27 am, Patrick Scheible <k...@zipcon.net> wrote:
> I have heard before
> that the French library community DOES sort by diacritical marks, as a
> less significant sorting element than the letters.

And that's perfectly acceptable. Sorting by diacritical marks *from
the end of a string*, so that a variance in diacritical marks later in
the string takes precedence over one earlier in the string, at least
seems bizarre (and I make no apologies for using this word in this
case) to an outsider.

John Savard

[Dan Espen]

I can't help but think some French Canadian came up with the scheme just
to cause problems, not because that was the prevailing scheme that was
in use.

--
Dan Espen

[John Levine]

>> I have heard before
>> that the French library community DOES sort by diacritical marks, as a
>> less significant sorting element than the letters.
>
>And that's perfectly acceptable. Sorting by diacritical marks *from
>the end of a string*, so that a variance in diacritical marks later in
>the string takes precedence over one earlier in the string, at least
>seems bizarre (and I make no apologies for using this word in this
>case) to an outsider.

It's totally dependent on the language. In French, � (o with a diaresis) is
a variant of a plain o, in German it's equivalent to oe, and in several
Scandinavian languages it's a letter unrelated to o that sorts at the end
of the alphabet.

Lengthy treatises have been written on the quirks of alphabetization.

--
Regards,
John Levine, jo...@iecc.com, Primary Perpetrator of "The Internet for Dummies",
Please consider the environment before reading this e-mail. http://jl.ly

[Ahem A Rivet's Shot]

Indeed it does, which is why it sticks in the mind as a good
example that alphabetic sorting gets complicated when internationalisation
is done correctly. It is one cause of apparent bloat, a lot of software
these days includes libraries with all these oddities coded in because once
you've seen a few examples like that you know you don't want to write or
maintain the alpha sorting code.

[Charlie Gibbs]

In article <86sj24g...@chai.my.domain>, k...@zipcon.net
(Patrick Scheible) writes:

> Peter Flass <Peter...@Yahoo.com> writes:
>
>> Unfortunately you can no longer get away with telling them "because
>> that's the way computers work, we have to do it that way." Used to
>> be a very effective response.

The response now is: "Because that's the way this software package
works." Or: "Because that's the way the latest update to this
software package works - isn't it _wonderful_?" Top it off with
a condescending look designed to make the user look totally stupid
should he/she have the temerity to ask why it couldn't just work
as well as the old version did.

> And made a whole lot of people hate computers.

A whole lot of people still do. But more and more have become
so worn down that they just give up and try to make do despite
whatever indignities the vendor subjects them to.

From what I've heard, it's sort of like life was in the
Soviet Union. Certainly the jokes are starting to sound
the same...

--
/~\ cgi...@kltpzyxm.invalid (Charlie Gibbs)
\ / I'm really at ac.dekanfrus if you read it the right way.
X Top-posted messages will probably be ignored. See RFC1855.
/ \ HTML will DEFINITELY be ignored. Join the ASCII ribbon campaign!

[Jean-Marc Bourguet]

I happen to have a dictionnary of 1865 where the rule is followed.

Yours,

--
Jean-Marc

[Dan Espen]

That's amazing.

Still waiting for some reason for a rule like that.
Until then, I'm going to assume that someone born before 1865 had a
perverse sense of humor.

--
Dan Espen

[Jaimie Vandenbergh]

On Sat, 04 May 2013 09:42:10 -0400, Dan Espen <des...@verizon.net>
wrote:

>Jean-Marc Bourguet <j...@bourguet.org> writes:
>
>> Dan Espen <des...@verizon.net> writes:
>>
>>> Quadibloc <jsa...@ecn.ab.ca> writes:
>>>
>>>> On May 3, 10:27 am, Patrick Scheible <k...@zipcon.net> wrote:
>>>>> I have heard before
>>>>> that the French library community DOES sort by diacritical marks, as a
>>>>> less significant sorting element than the letters.
>>>>
>>>> And that's perfectly acceptable. Sorting by diacritical marks *from
>>>> the end of a string*, so that a variance in diacritical marks later in
>>>> the string takes precedence over one earlier in the string, at least
>>>> seems bizarre (and I make no apologies for using this word in this
>>>> case) to an outsider.
>>>
>>> I can't help but think some French Canadian came up with the scheme just
>>> to cause problems, not because that was the prevailing scheme that was
>>> in use.
>>
>> I happen to have a dictionnary of 1865 where the rule is followed.
>>
>> Yours,
>
>That's amazing.

It's not at all. You're being embarrassingly provincial.

>Still waiting for some reason for a rule like that.

A number of languages that use accent marks and ligatures treat the
resulting letter not as (eg) "an A with a circle over it" but as a
different letter from A entirely. Thus it gets its own sorting rules.

This should be no harder to comprehend than a local rule to sort
numbers after (or before) letters.

>Until then, I'm going to assume that someone born before 1865 had a
>perverse sense of humor.

ASCII - and specifically America - is not the world.

Cheers - Jaimie
--
There's no limit to the amount of work someone can do,
provided it's not the work they're supposed to be doing.

[Dan Espen]

I think you better take a closer look.

It's the switching direction part,
something only done in Canadian French,
that's perverse.

How many other languages does that occur in?
Apparently not French.

--
Dan Espen

[Peter Flass]

Yes, but sorting characters left to right until you encounter some
character with a doohickey on it and then switching to sort right to
left *is* bizarre. Otherwise, yes, different languages have different
rules for sorting characters with diacritical marks - for example German
sorts a and ä the same.


--
Pete

[jmfbahciv]

Maybe it was a first attempt at establishing sorting rules. When was
the first dictionary written? I remember reading about the history
of the OED but that's not the quite the same.

/BAH

>

[Patrick Scheible]

The Academie Francaise published its first dictionary of French in 1694,
says Wikipedia. The first English dictionaries were about the same
time, although until Samuel Johnson's of 1755 they were not complete
guides to the languages, but just lists of hard words.

Dictionaries of other languages go back to ancient times.

-- Patrick

[Jaimie Vandenbergh]

On Sat, 04 May 2013 20:31:09 -0400, Dan Espen <des...@verizon.net>

Yeah, to be fair that is pretty bizarre. In twelve years of working
with sorting name+address data (telephone directories biz) it's only
come up that once. Other peculiar sorts - ordering by first name
unless surname is not Icelandic, for example - are pretty common; and
everywhere has different orderings for diacritics.

Cheers - Jaimie
--
Actually, the Singularity seems rather useful in the entire work avoidance
field. "I _could_ write up that report now but if I put it off, I may well
become a weakly godlike entity, at which point not only will I be able to
type faster but my comments will be more on-target." - James Nicoll

[hanc...@bbs.cpcn.com]

On May 3, 12:30 pm, Patrick Scheible <k...@zipcon.net> wrote:

Good question. FWIW, the movie was in IIRC 1954, and the interchange
at hand was a small one. Even in the 1980s there could be light
traffic at the smaller interchanges, light enough that a toll taker
was able to 'entertain' a trucker in her booth for a while until
getting caught. In the 1950s, I could see traffic being extremely
light at some of the rural interchanges.

Also, the turnpike was relatively new back then. Back then, the newly
built turnpikes had a very pro-active public realtions attitude in
order to get motorists to be willing to pay a toll for something they
could do on other roads for free. They issued quite a bit of
literature, films, and whatnot (along with oil companies and car
makers) to encourage use*. Accordingly, in that spirit, I could see
them helping out with such a request, especially since it was from a
nearby major company and a light interchange.

*Master builder Robert Moses' Triborough agency was noted for
producing very high quality printed materials to promote its
facilities.

[Rod Speed]

"jmfbahciv" <See....@aol.com> wrote in message
news:PM0004DBF...@ac8101d7.ipt.aol.com...

Whoops, went to check wikipedia on that and it
looks like someone has fucked that entry over
http://en.wikipedia.org/wiki/Dictionary#your_an_idiot_if_your_reading_Wikipedia_to_find_information_for_any_projects_or_reports_but_any_way_lets_start_withHistory

It appears to be saying something like 2300BCE but
I haven't bothered to check the state of the article
before it got fucked over.

> I remember reading about the history of
> the OED but that's not the quite the same.

That wasn’t even the first one for english.

[Quadibloc]

On May 5, 10:50 am, Jaimie Vandenbergh <jai...@sometimes.sessile.org>
wrote:

> On Sat, 04 May 2013 20:31:09 -0400, Dan Espen <des...@verizon.net>
> wrote:

> >It's the switching direction part,
> >something only done in Canadian French,
> >that's perverse.
>
> Yeah, to be fair that is pretty bizarre. In twelve years of working
> with sorting name+address data (telephone directories biz) it's only
> come up that once. Other peculiar sorts - ordering by first name
> unless surname is not Icelandic, for example - are pretty common; and
> everywhere has different orderings for diacritics.

From
http://blogs.msdn.com/b/michkap/archive/2004/12/31/344739.aspx
apparently it is done in France, in Quebec, and everywhere the French
language is used!!

John Savard

[Quadibloc]

> >It's the switching direction part,
> >something only done in Canadian French,
> >that's perverse.

http://userguide.icu-project.org/collation/concepts

claims that it is _now_ only done in Canadian French - other locales
just gave it up very recently, apparently.

John Savard

[Dan Espen]

Seems even some French people agree.
Just say no!

--
Dan Espen

[Peter Flass]

Some day they'll figure out what a collating sequence is.


--
Pete

[jmfbahciv]

In France that wouldn't be a valid excuse. I was told that they
purge words which aren't French every five years. The purpose is
to keep the language pure and not allow other lanugages' terms
creep into the vocabulary. They must have had to work very hard
to keep DEC terms out of their vocabulary. I often wondered
what they did with IBM's.

/BAH

[jmfbahciv]

Do you know how they were organized?

/BAH

[Patrick Scheible]

Partially.

The Urra=hubullu, a Sumerian and Akkadian lexicon from circa 2300 BC, is
by topic. (The title is from the first entry, for interest-bearing
debt).

The Erya, a Chinese dictionary/encyclopedia from about the 3rd century
BC, is organized into chapters by subject and most are subdivided by
radicals, though some have other arrangements.

I'm not finding a description online of Apollonius the Sophist's lexicon
of words found in Homer, so I'm having a reprinted edition sent.

-- Patrick

[Charlie Gibbs]

In article <icmws9d...@home.home>, des...@verizon.net (Dan Espen)
writes:

I think that should be: "Just say non!"

[Dan Espen]

"Charlie Gibbs" <cgi...@kltpzyxm.invalid> writes:

> In article <icmws9d...@home.home>, des...@verizon.net (Dan Espen)
> writes:
>
>> Quadibloc <jsa...@ecn.ab.ca> writes:
>>
>>>>> It's the switching direction part,
>>>>> something only done in Canadian French,
>>>>> that's perverse.
>>>
>>> http://userguide.icu-project.org/collation/concepts
>>>
>>> claims that it is _now_ only done in Canadian French - other locales
>>> just gave it up very recently, apparently.
>>
>> Seems even some French people agree.
>> Just say no!
>
> I think that should be: "Just say non!"

Google assures me it should be "juste dire non!".

--
Dan Espen

[Peter Flass]

I think some of the Latin ones were just lists of problem words, as
someone has already described. What's interesting is the the "errors,"
mispronunciations, etc. give an indication of the different ways the
language was evolving.


--
Pete

[Rod Speed]

"jmfbahciv" <See....@aol.com> wrote in message

news:PM0004DC0...@aca30337.ipt.aol.com...

> Peter Flass wrote:
>> On 5/5/2013 6:19 PM, Quadibloc wrote:
>>> On May 5, 10:50 am, Jaimie Vandenbergh <jai...@sometimes.sessile.org>
>>> wrote:
>>>> On Sat, 04 May 2013 20:31:09 -0400, Dan Espen <des...@verizon.net>
>>>> wrote:
>>>
>>>>> It's the switching direction part,
>>>>> something only done in Canadian French,
>>>>> that's perverse.
>>>>
>>>> Yeah, to be fair that is pretty bizarre. In twelve years of working
>>>> with sorting name+address data (telephone directories biz) it's only
>>>> come up that once. Other peculiar sorts - ordering by first name
>>>> unless surname is not Icelandic, for example - are pretty common; and
>>>> everywhere has different orderings for diacritics.
>>>
>>> From
>>> http://blogs.msdn.com/b/michkap/archive/2004/12/31/344739.aspx
>>> apparently it is done in France, in Quebec, and everywhere the French
>>> language is used!!
>>>
>>
>> Some day they'll figure out what a collating sequence is.

> In France that wouldn't be a valid excuse. I was
> told that they purge words which aren't French

Yes, they do attempt stupiditys like that.

> every five years.

It isnt done like that.

> The purpose is to keep the language pure and not
> allow other lanugages' terms creep into the vocabulary.

Yes.

> They must have had to work very hard
> to keep DEC terms out of their vocabulary.
> I often wondered what they did with IBM's.

Not sure what they do about technical terms like that.

[Andrew Swallow]

On 06/05/2013 14:15, jmfbahciv wrote:
{snip}

> In France that wouldn't be a valid excuse. I was told that they
> purge words which aren't French every five years. The purpose is
> to keep the language pure and not allow other lanugages' terms
> creep into the vocabulary. They must have had to work very hard
> to keep DEC terms out of their vocabulary. I often wondered
> what they did with IBM's.
>
> /BAH
>

You don't think that the committee is going to dirty its hands dealing
with technical publications do you?

The only technical words they will cover are the ones that appear in
government regulations.

Andrew Swallow

[Gene Wirchenko]

On Mon, 06 May 2013 15:03:21 -0400, Peter Flass
<Peter...@Yahoo.com> wrote:

[snip]

>I think some of the Latin ones were just lists of problem words, as
>someone has already described. What's interesting is the the "errors,"
>mispronunciations, etc. give an indication of the different ways the
>language was evolving.

I expect that "supersede" in English is probably from one of
those errors.

Sincerely,

Gene Wirchenko

[Alfred Falk]

Gene Wirchenko <ge...@telus.net> wrote in
news:favgo8hgfa9jqpcgu...@4ax.com:

How so? Latin root is "sedere", to sit, methinks.

[Jean-Marc Bourguet]

jmfbahciv <See....@aol.com> writes:

> In France that wouldn't be a valid excuse. I was told that they
> purge words which aren't French every five years.

Who "they"? They purge words from what?

Yours,

--
Jean-Marc

[jmfbahciv]

When I was in the RUNOFF group, people would write the word
as superceed.

/BAH

[Dan Espen]

This seemed like another case of an unidentified mythical enemy but
Google says otherwise. I see references to:

French government’s general commission on terminology and neology.

This article:

http://online.wsj.com/article/SB125544523318682497.html

Says:

In 1994, the French government passed laws to ensure that all
advertisements, work-contracts and government documentation were in
French. The General Delegation was charged with overseeing the
creation of new French terms.
...
Each of France's government ministries has at least one terminology
committee attached to it. The job of the people on the committee is to
spot new English words and create and define French alternatives
before the English version catches on.

It's nuts!


--
Dan Espen

[Stan Barr]

Webster's has it coming from the Old French "superseder" which
probably does have its roots in Latin.

--
Cheers,
Stan Barr plan.b .at. dsl .dot. pipex .dot. com

The future was never like this!

[Peter Flass]

When a language starts having to be supervised by policemen, it's on the
way out.

--
Pete

[Peter Flass]

On 5/8/2013 11:20 AM, Stan Barr wrote:
> On Tue, 7 May 2013 18:27:56 +0000 (UTC), Alfred Falk
> <fa...@arc.REMOVE.ab.ca> wrote:
>> Gene Wirchenko <ge...@telus.net> wrote in
>> news:favgo8hgfa9jqpcgu...@4ax.com:
>>
>>> On Mon, 06 May 2013 15:03:21 -0400, Peter Flass
>>> <Peter...@Yahoo.com> wrote:
>>>
>>> [snip]
>>>
>>>> I think some of the Latin ones were just lists of problem words, as
>>>> someone has already described. What's interesting is the the "errors,"
>>>> mispronunciations, etc. give an indication of the different ways the
>>>> language was evolving.
>>>
>>> I expect that "supersede" in English is probably from one of
>>> those errors.
>>
>> How so? Latin root is "sedere", to sit, methinks.
>
> Webster's has it coming from the Old French "superseder" which
> probably does have its roots in Latin.
>

"Superseder" sounds like a real blowout Passover dinner.


--
Pete

[Anne & Lynn Wheeler]

Peter Flass <Peter...@Yahoo.com> writes:
> When a language starts having to be supervised by policemen, it's on
> the way out.

early 70s, I was in large business meeting in paris ... where somebody
from la guade was making presentation. somebody from paris kept
interrupting him to correct his pronunciation.

--
virtualization experience starting Jan1968, online at home since Mar1970

[Rod Speed]

"Jean-Marc Bourguet" <j...@bourguet.org> wrote in message
news:87ehdhd...@news.bourguet.org...

> jmfbahciv <See....@aol.com> writes:
>
>> In France that wouldn't be a valid excuse. I was told that they
>> purge words which aren't French every five years.

> Who "they"?

Acad�mie fran�aise

> They purge words from what?

Their dictionary.

http://en.wikipedia.org/wiki/Acad%C3%A9mie_fran%C3%A7aise#Role_as_authority_on_the_French_language

[Patrick Scheible]

Dan Espen <des...@verizon.net> writes:

Dan Espen <des...@verizon.net> writes:

It's the Academie Francaise given the force of law.

They see it as preserving an essential part of French culture. But to
English speakers it seems like nuts, because English has always
enthusiastically adopted foreign words and phrases into its vocabulary.

-- Patrick

[Dave Garland]

On 5/8/2013 12:35 PM, Patrick Scheible wrote:

> Dan Espen <des...@verizon.net> writes:
>

>> This article:
>>
>> http://online.wsj.com/article/SB125544523318682497.html
>>
>> Says:
>>
>> In 1994, the French government passed laws to ensure that all
>> advertisements, work-contracts and government documentation were in
>> French. The General Delegation was charged with overseeing the
>> creation of new French terms.
>> ...
>> Each of France's government ministries has at least one terminology
>> committee attached to it. The job of the people on the committee is to
>> spot new English words and create and define French alternatives
>> before the English version catches on.
>>
>> It's nuts!
>
> It's the Academie Francaise given the force of law.
>
> They see it as preserving an essential part of French culture. But to
> English speakers it seems like nuts, because English has always
> enthusiastically adopted foreign words and phrases into its vocabulary.
>

"The problem with defending the purity of the English language is that
English is about as pure as a cribhouse whore. We don't just borrow
words; on occasion, English has pursued other languages down alleyways
to beat them unconscious and riffle their pockets for new vocabulary."
-James Nicoll

[sdrat]

"Peter Flass" <Peter...@Yahoo.com> wrote in message
news:kmdqsl$cau$1...@dont-email.me...

It isn't supervised by policemen.
http://en.wikipedia.org/wiki/Acad%C3%A9mie_fran%C3%A7aise#Role_as_authority_on_the_French_language

[hanc...@bbs.cpcn.com]

On May 8, 11:55 am, Anne & Lynn Wheeler <l...@garlic.com> wrote:

> early 70s, I was in large business meeting in paris ... where somebody
> from la guade was making presentation. somebody from paris kept
> interrupting him to correct his pronunciation.

A number of British, Australian, and New Zealand actors have taken
roles in the US, devoid of any accent. This would include the
girlfriend on "Chuck", Lucy Lawless on Xena, the doctor on "House",
and Emma Watson ("Wallflower").

One time that did _not_ work out was when they had Sidney Greenstreet
playing a Southern businessman. He was a great actor, but some things
were too much of a stretch. (He's in a movie running on TCM now on
demand, "Verdict", playing an inspector from Scotland Yard.)

[Peter Flass]

From what I hear of German form TV shows (thanks, Internet!) they've
adopted an awful lot of English. "OK" seems to be the most common word
of agreement.


--
Pete

[Peter Flass]

Simon Baker (Patrick Jayne on the "Mentalist") is another. I was amazed
to hear him speak in his natural accent rather than in character.

I'd like to know how they do it and how long it takes, their 'Merican
accents are flawless.

--
Pete

[Gene Wirchenko]

On Tue, 7 May 2013 18:27:56 +0000 (UTC), Alfred Falk
<fa...@arc.REMOVE.ab.ca> wrote:

Maybe, it is a bit of folk etymology. We do have "-cede"s and
"-ceed"s in English, but "supersede" is the only "-sede" that I can
think of.

Sincerely,

Gene Wirchenko

[Quadibloc]

On May 8, 9:37 am, Peter Flass <Peter_Fl...@Yahoo.com> wrote:

> When a language starts having to be supervised by policemen, it's on the
> way out.

While the government commission only started up in 1994, the Academie
Francaise had been around for quite a bit longer - and I've read
somewhere that a few other European countries have similar
institutions, even if the French one gets the most noticed.

At the moment, it does indeed look like English is poised to take over
the world. But this could be just a phase - which will end when they
finally get Google Translate working correctly.

John Savard

[John Levine]

>> When a language starts having to be supervised by policemen, it's on the
>> way out.

The Acad�mie Fran�aise was founded in 1635, and has been operating
ever since except for 1793-1803 after the revolution. That's quite a
lengthy exit.

I think you will find that it is more common than not for countries to
have something similar to define what their language is. English is
unusual in having nobody in charge and nobody really caring about it.

--
Regards,
John Levine, jo...@iecc.com, Primary Perpetrator of "The Internet for Dummies",
Please consider the environment before reading this e-mail. http://jl.ly

[Dan Espen]

John Levine <jo...@iecc.com> writes:

>>> When a language starts having to be supervised by policemen, it's on the
>>> way out.
>

> The Acadmie Franaise was founded in 1635, and has been operating
^^^^ not really working out...

> ever since except for 1793-1803 after the revolution. That's quite a
> lengthy exit.
>
> I think you will find that it is more common than not for countries to
> have something similar to define what their language is. English is
> unusual in having nobody in charge and nobody really caring about it.

Cite?

Lot of discussion here:

http://en.wikipedia.org/wiki/Linguistic_purism

but it's not that specific on how often definition is a government
function.

--
Dan Espen

[Charlie Gibbs]

In article <kme3fh$13s$1...@dont-email.me>, dave.g...@wizinfo.com

I've heard English described as a "kleptolect".

[Andrew Swallow]

Policemen do not make rules, they enforce rules made by other people.
Both groups have been shown to exist for the French language above.

Andrew Swallow

[sdrat]

"Andrew Swallow" <am.sw...@btinternet.com> wrote in message
news:Ru-dnaNcD6xaiBbM...@bt.com...

Yes, but that does not happen with the french language
and as that article says, rulings by the Académie française
do in fact just get ignored at times. There is no supervision
by policemen either.

> Both groups have been shown to exist for the French language above.

No, there is no supervision by policemen.

The most there is is the Académie française attempting to rule on what is
acceptable.

[Gene Wirchenko]

On 08 May 13 19:42:08 -0800, "Charlie Gibbs" <cgi...@kltpzyxm.invalid>
wrote:

[snip]

>I've heard English described as a "kleptolect".

It is especially appropriate considering that both parts of the
word ("klepto-" and "-lect") come from Greek.

Sincerely,

Gene Wirchenko

[Patrick Scheible]

Changing accents is something actors learn to do.

-- Patrick

[Nick Spalding]

Gene Wirchenko wrote, in <qgelo815qkv0m5qm0...@4ax.com>
on Wed, 08 May 2013 13:50:04 -0700:

In my BrE vocabulary it is "supercede" but I see that Agent's English
(International) spell checker accepts either.
--
Nick Spalding

[Jean-Marc Bourguet]

"Rod Speed" <rod.sp...@gmail.com> writes:

> "Jean-Marc Bourguet" <j...@bourguet.org> wrote in message
> news:87ehdhd...@news.bourguet.org...
>> jmfbahciv <See....@aol.com> writes:
>>
>>> In France that wouldn't be a valid excuse. I was told that they
>>> purge words which aren't French every five years.
>
>> Who "they"?
>
> Acad�mie fran�aise
>
>> They purge words from what?
>
> Their dictionary.

That's what I first though, but considering that

1/ their dictionnary has 50 years or more between editions, not 5
years. (And with such a slow cycle, keeping out the latest trendy words
seems wize, they probably won't be used at all for the major part of the
life time of the dictionnary)

2/ Purge for me imply that those words were first put in the dictionnary.

it didn't seem the correct guess. Still doesn't.

Yours,

--
Jean-Marc

[Jean-Marc Bourguet]

Dan Espen <des...@verizon.net> writes:

> Jean-Marc Bourguet <j...@bourguet.org> writes:
>
>> jmfbahciv <See....@aol.com> writes:
>>
>>> In France that wouldn't be a valid excuse. I was told that they
>>> purge words which aren't French every five years.
>>
>> Who "they"? They purge words from what?
>>
>> Yours,
>
> This seemed like another case of an unidentified mythical enemy but
> Google says otherwise. I see references to:
>
> French government’s general commission on terminology and neology.
>
> This article:
>
> http://online.wsj.com/article/SB125544523318682497.html
>
> Says:
>
> In 1994, the French government passed laws to ensure that all
> advertisements, work-contracts and government documentation were in
> French.

The goal isn't to avoid specific terms but to avoid getting user level
documentation provided only in languages not understood by customers.
(Nowadays, the translation seems to be done by translate.google.com...)

> The General Delegation was charged with overseeing the
> creation of new French terms.
> ...
> Each of France's government ministries has at least one terminology
> committee attached to it. The job of the people on the committee is to
> spot new English words and create and define French alternatives
> before the English version catches on.

With or without that law, some more french sounding alternatives are
created and used, usually several for each concept. The effect of those
committees, if they have one something I'm not sure, is that government
issue documents should be using the same one, avoiding the issues "is there
are difference of meaning between these two different words in those two
different documents?" and "that same word in those two documents seem to
mean something different". They have little practical effect on common
usage.


AFAIK, the Real Academia Espagñola and the Nederlandse Taalunie have more
influence on the practical usage of the languages they supervise than the
Académie française or those committees have on french.

Yours,

--
Jean-Marc

[Jean-Marc Bourguet]

Peter Flass <Peter...@Yahoo.com> writes:

> When a language starts having to be supervised by policemen, it's on the
> way out.

Remember to never split an infinitive. The passive voice should never
be used. Do not put statements in the negative form. Verbs have to
agree with their subjects. Proofread carefully to see if you words
out. If you reread your work, you can find on rereading a great deal
of repetition can be avoided by rereading and editing. A writer must
not shift your point of view. And don't start a sentence with a
conjunction. (Remember, too, a preposition is a terrible word to end a
sentence with.) Don't overuse exclamation marks!! Place pronouns as
close as possible, especially in long sentences, as of 10 or more
words, to their antecedents. Writing carefully, dangling participles
must be avoided. If any word is improper at the end of a sentence, a
linking verb is. Take the bull by the hand and avoid mixing
metaphors. Avoid trendy locutions that sound flaky. Everyone should
be careful to use a singular pronoun with singular nouns in their
writing. Always pick on the correct idiom. The adverb always follows
the verb. Last but not least, avoid cliches like the plague; seek
viable alternatives.


Seems you have the same kind of rule makers and (self appointed) policemen
for English as well. They have about the same influence for French as for
English (not null but far from the one they would like).

Yours,

--
Jean-Marc

[Peter Flass]

I know German was updated a few years ago. It required a multi-national
commission. As I recall they mostly tinkered with the spelling.

--
Pete

[Peter Flass]

A dictionary is supposed to reflect the language a actually
spoken/written, not as someone thinks it should be. It's worth putting
in trendy words, *especially* when they might not be around in 25 or 50
years. Someone 100 years from now will come across "quwezzt" and want
to know what it meant.

--
Pete

[Peter Flass]

That's the thing about technical terms, *especially* if they're not
otherwise a common word in your language. It has only one meaning and
doesn't drag a lot of baggage along with it. That's why plants and
animals have Latin names. They naturally have at least one and
frequently more names in each language, and the same name occasionally
refers to several different things, but the Latin term can mean only one
thing.


--
Pete

[Jean-Marc Bourguet]

Peter Flass <Peter...@Yahoo.com> writes:

> A dictionary is supposed to reflect the language a actually spoken/written,
> not as someone thinks it should be.

If you want a dictionnary reflecting the language at a given time, use one
made by lexicographers, not by a bunch of persons which coopt the successor
of those who die.

> It's worth putting in trendy words, *especially* when they might not be
> around in 25 or 50 years. Someone 100 years from now will come across
> "quwezzt" and want to know what it meant.

It is worth when you update your dictionnary more often than once every 50
years. Dictionnaries which have a yearly edition get words in (and out)
each time.

Yours,

--
Jean-Marc