typo in frame.el

typo in frame.el

[Hiroshi Fujishima]

I found typo in frame.el.

--- frame.el	30 12月 2003 04:17:24 +0900	1.206
+++ frame.el	09  4月 2004 19:18:21 +0900	
@@ -1248,7 +1248,7 @@
 
 (defun blink-cursor-mode (arg)
   "Toggle blinking cursor mode.
-With a numeric argument, turn blinking cursor mode on iff ARG is positive.
+With a numeric argument, turn blinking cursor mode on if ARG is positive.
 When blinking cursor mode is enabled, the cursor of the selected
 window blinks.

-- 
Hiroshi Fujishima

Re: typo in frame.el

[David Kastrup]

Hiroshi Fujishima <pooh@nature.tsukuba.ac.jp> writes:

> I found typo in frame.el.
> 
> --- frame.el	30 12月 2003 04:17:24 +0900	1.206
> +++ frame.el	09  4月 2004 19:18:21 +0900	
> @@ -1248,7 +1248,7 @@
>  
>  (defun blink-cursor-mode (arg)
>    "Toggle blinking cursor mode.
> -With a numeric argument, turn blinking cursor mode on iff ARG is positive.
> +With a numeric argument, turn blinking cursor mode on if ARG is positive.
>  When blinking cursor mode is enabled, the cursor of the selected
>  window blinks.

That's not a typo.  That is a rather common shorthand for "if and only
if".

-- 
David Kastrup, Kriemhildstr. 15, 44793 Bochum

Re: typo in frame.el

[Alex Schroeder]

Hiroshi Fujishima <pooh@nature.tsukuba.ac.jp> writes:

> -With a numeric argument, turn blinking cursor mode on iff ARG is positive.
> +With a numeric argument, turn blinking cursor mode on if ARG is positive.

For mathematicians, "iff" is an abbreviation for "if and only
if"...  So it is not a typo, I suppose.  I think that should avoid the
use of such abbreviations in Emacs documentation, however, since not
all our users come from an English/science background.

Alex.
-- 
.O.  http://www.emacswiki.org/alex/
..O  Schroeder's fourth law:
OOO  None of your friends and coworkers share your taste in music.

RE: typo in frame.el

[Drew Adams]

Doc strings, at least their first lines, need to be short and clear, so
"iff" is in fact helpful - but only if the convention is understood.

Instead of not taking advantage of the convention, why don't we include an
explanation of it in the general doc: add a line stating what "iff" means to
the Emacs (or Elisp) doc?

At the very worst, someone unacquainted with the meaning of "iff" will
likely interpret it as "if" (and as an benign typo), so there is no harm
done in using it: (the most important) half of its meaning is always
conveyed.

Another possibility is to use "<=>" (and "=>" for "implies"), but I suppose
that assumes some familiarity with symbolic logic or math.

 - Drew

-----Original Message-----
From: emacs-devel-bounces+drew.adams=oracle.com@gnu.org
[mailto:emacs-devel-bounces+drew.adams=oracle.com@gnu.org]On Behalf Of
Alex Schroeder
Sent: Friday, April 09, 2004 11:58 AM
To: Hiroshi Fujishima
Cc: emacs-devel@gnu.org
Subject: Re: typo in frame.el


Hiroshi Fujishima <pooh@nature.tsukuba.ac.jp> writes:

> -With a numeric argument, turn blinking cursor mode on iff ARG is
positive.
> +With a numeric argument, turn blinking cursor mode on if ARG is positive.

For mathematicians, "iff" is an abbreviation for "if and only
if"...  So it is not a typo, I suppose.  I think that should avoid the
use of such abbreviations in Emacs documentation, however, since not
all our users come from an English/science background.

Alex.
--
.O.  http://www.emacswiki.org/alex/
..O  Schroeder's fourth law:
OOO  None of your friends and coworkers share your taste in music.



_______________________________________________
Emacs-devel mailing list
Emacs-devel@gnu.org
http://mail.gnu.org/mailman/listinfo/emacs-devel

Re: typo in frame.el

[Simon Josefsson]

Alex Schroeder <alex@emacswiki.org> writes:

> Hiroshi Fujishima <pooh@nature.tsukuba.ac.jp> writes:
>
>> -With a numeric argument, turn blinking cursor mode on iff ARG is positive.
>> +With a numeric argument, turn blinking cursor mode on if ARG is positive.
>
> For mathematicians, "iff" is an abbreviation for "if and only
> if"...  So it is not a typo, I suppose.  I think that should avoid the
> use of such abbreviations in Emacs documentation, however, since not
> all our users come from an English/science background.

I prefer "precisely when" instead of "if and only if".  Sometimes,
even the "precisely" can be omitted.  Less technobabbly.  Alas,
changing "iff" into "when" can make a documentation string line too
long.  Confusion regarding "iff" is a reoccurring theme, so it might
make sense to do something about it.

(Kudos to my algebra professor, who introduced me to p.w., although
only for Swedish.  But I think it works as well in English.  Any
native speaker who disagree?)

RE: typo in frame.el

[Drew Adams]

How about we use "<=>", conventionally, in doc strings, but we also explain
this convention in the Emacs manual, for those who have never seen it? If we
adopt such a convention, then we should perhaps also conventionally use "=>"
for "implies" ("only if") and "<=" for "if".

(If you don't enjoy semantic nitpicking, hit the delete key now.)

In "precisely when", the key word, which makes it equivalent to "iff", is
"precisely". If you drop it, then you're back to just "if": "when", by
itself, means just "if" - that is:

  "a when b" <=> "a if b" <=> "b only if a" <=> "if b then a"

I think "iff" is much more common, in English technical writing, than
"precisely when". ("<=>" is more common than "iff" in math: it is not tied
to a natural language.)

Also, it is fairly common in programming to use "when" to indicate one or
both of these (as opposed to just logical implication):

 - temporal relation ("at the same time as")
 - condition for a temporal sequence of actions/events

For example, in Common Lisp, "when" is used stylistically to introduce a
conditional sequence of actions (with side effects).

 - Drew

-----Original Message-----
From: emacs-devel-bounces+drew.adams=oracle.com@gnu.org
[mailto:emacs-devel-bounces+drew.adams=oracle.com@gnu.org]On Behalf Of
Simon Josefsson
Sent: Friday, April 09, 2004 1:22 PM
To: emacs-devel@gnu.org
Subject: Re: typo in frame.el


Alex Schroeder <alex@emacswiki.org> writes:

> Hiroshi Fujishima <pooh@nature.tsukuba.ac.jp> writes:
>
>> -With a numeric argument, turn blinking cursor mode on iff ARG is
positive.
>> +With a numeric argument, turn blinking cursor mode on if ARG is
positive.
>
> For mathematicians, "iff" is an abbreviation for "if and only
> if"...  So it is not a typo, I suppose.  I think that should avoid the
> use of such abbreviations in Emacs documentation, however, since not
> all our users come from an English/science background.

I prefer "precisely when" instead of "if and only if".  Sometimes,
even the "precisely" can be omitted.  Less technobabbly.  Alas,
changing "iff" into "when" can make a documentation string line too
long.  Confusion regarding "iff" is a reoccurring theme, so it might
make sense to do something about it.

(Kudos to my algebra professor, who introduced me to p.w., although
only for Swedish.  But I think it works as well in English.  Any
native speaker who disagree?)



_______________________________________________
Emacs-devel mailing list
Emacs-devel@gnu.org
http://mail.gnu.org/mailman/listinfo/emacs-devel

Re: typo in frame.el

[David Kastrup]

"Drew Adams" <drew.adams@oracle.com> writes:

> How about we use "<=>", conventionally, in doc strings, but we also
> explain this convention in the Emacs manual, for those who have
> never seen it? If we adopt such a convention, then we should perhaps
> also conventionally use "=>" for "implies" ("only if") and "<=" for
> "if".

I object.  Really.  This would carry the message that Emacs is
intended for mathematicians and/or scientists as main audience.  Apart
from which those symbols are just ASCII art representations from what
they are supposed to be.

-- 
David Kastrup, Kriemhildstr. 15, 44793 Bochum

Re: typo in frame.el

[Simon Josefsson]

David Kastrup <dak@gnu.org> writes:

> "Drew Adams" <drew.adams@oracle.com> writes:
>
>> How about we use "<=>", conventionally, in doc strings, but we also
>> explain this convention in the Emacs manual, for those who have
>> never seen it? If we adopt such a convention, then we should perhaps
>> also conventionally use "=>" for "implies" ("only if") and "<=" for
>> "if".
>
> I object.  Really.  This would carry the message that Emacs is
> intended for mathematicians and/or scientists as main audience.  Apart
> from which those symbols are just ASCII art representations from what
> they are supposed to be.

I agree.  The problem seem to be to find a suitable English expression
that embodies the same meaning.  "Iff" is probably not a good idea,
but I'm not sure expanding it to "if and only if" is enough.  OTOH,
perhaps if the expression is only used in very technical situations,
it would be OK.  Surely someone who program elisp understand
it (although might not understand "iff").

RE: typo in frame.el

[Drew Adams]

Hmmm...  I suppose so. But two silly counter-arguments, then I'll rest:

1) The unwashed masses seem to have digested ":)" OK... And they can handle
magic-eye SIRDS dinosaurs OK. Bringing them up to speed on an ASCII arrow
that pretty-much shows what it says should be a piece of cake, non?

(Then again, at least the Murrikens think Mr. Hussein was in NY on 9/11.)

2) Most emacs users are, and have been, (dare-I-say-it?) programmers. I
doubt that "<=>" is very foreign to most of them. If foreign, I doubt they
would have difficulty picking it up. And emacsers already speak "C-" and
"M-". How hard is this?

Yeah, I know, Emacs is not *aimed* only at programmers, and I would love to
see the Emacs word-processing functionality (and more!) that RMS dreams of.
But, so far...

Cheers & "Uncle!",

   Drew

-----Original Message-----
From: dak@lola.goethe.zz [mailto:dak@lola.goethe.zz]On Behalf Of David
Kastrup
Sent: Friday, April 09, 2004 2:11 PM
To: Drew Adams
Cc: Simon Josefsson; emacs-devel@gnu.org
Subject: Re: typo in frame.el


"Drew Adams" <drew.adams@oracle.com> writes:

> How about we use "<=>", conventionally, in doc strings, but we also
> explain this convention in the Emacs manual, for those who have
> never seen it? If we adopt such a convention, then we should perhaps
> also conventionally use "=>" for "implies" ("only if") and "<=" for
> "if".

I object.  Really.  This would carry the message that Emacs is
intended for mathematicians and/or scientists as main audience.  Apart
from which those symbols are just ASCII art representations from what
they are supposed to be.

--
David Kastrup, Kriemhildstr. 15, 44793 Bochum

Re: typo in frame.el

[David Kastrup]

"Drew Adams" <drew.adams@oracle.com> writes:

> Hmmm...  I suppose so. But two silly counter-arguments, then I'll rest:
> 
> 1) The unwashed masses seem to have digested ":)" OK...

"Digested"?  I'd say they created it.  I doubt that hackers need
explicit warning labels for nonobvious humor.

> And they can handle magic-eye SIRDS dinosaurs OK. Bringing them up
> to speed on an ASCII arrow that pretty-much shows what it says
> should be a piece of cake, non?

"<=>" is far from being anything like an arrow.  It is probably a
long time since you last played with a bow.  If anything, and arrow
is something like "-------->".  The symbol is shortened beyond
recognition, and also it has a doubled shaft.  The doubled shaft
carries the mathematical meaning of "for all variables" or "always".
a->b is a statement with a truth value.  a=>b states that for all
possible variables, this statement is true.

a <=> b is a pretty complex concept.  And I don't agree with
"pretty-much shows what is says".  If a shoots an arrow at b, I fail
to see how this makes b at least as truthful as a.  Even if you
consider an arrow a treacherous weapon, the truthfulness of b does
not depend on the treachery of a.

> 2) Most emacs users are, and have been, (dare-I-say-it?)
> programmers. I doubt that "<=>" is very foreign to most of them. If
> foreign, I doubt they would have difficulty picking it up. And
> emacsers already speak "C-" and "M-". How hard is this?

It's nothing at all related to "C-" and "M-".  The latter are not
mathematician-specific, but Emacs specific lingo.  They are explained
right on Emacs' splash screen.  And they serve an important purpose:
that of talking about key bindings that have no other equally short
name.

Do you want to put on Emacs' splash screen "a <=> b  means  a if and
only if b"?  Really?

> Yeah, I know, Emacs is not *aimed* only at programmers, and I would
> love to see the Emacs word-processing functionality (and more!) that
> RMS dreams of.

Really, I don't know why I am working on getting AUCTeX and
preview-latex to create easy interfaces to WYSIWYG LaTeX creation.  I
am working hard to provide an increasingly compelling environment also
for non-programmers for text processing.

I just fail to see what purpose <=> would serve that could not be done
otherwise.  Even "iff" is wagonloads better.  People that don't
understand that terminology are probably prone to not interpreting
"if" in the mathematical way, anyhow.  So if they consider it as a
typo and apply the everyday meaning of "if" to it instead of the
mathematical meaning, they are pretty much fine.

-- 
David Kastrup, Kriemhildstr. 15, 44793 Bochum

Re: typo in frame.el

[Alan Shutko]

"Drew Adams" <drew.adams@oracle.com> writes:

> 2) Most emacs users are, and have been, (dare-I-say-it?) programmers. I
> doubt that "<=>" is very foreign to most of them. 

FWIW I've been through several symbolic logic classes (two semesters
on just symbolic logic, as well as the usual intros in about three or
four other classes) and I didn't pick up on that meaning of "<=>"
until I saw it explained.  Whereas "iff" has been perfectly clear to
me every time I've seen it.

Maybe it's the double lines.  Maybe it's the fact that it's not
connected.  Maybe it's that it looks like a little UFO in an ascii
game.  

I'd say that "<=>" is a huge step backwards in understandability.
Why should we pick that particular ASCII rendition of a symbol from
symbolic logic, especially since I'd wager that's not the only symbol
used for iff even in symbolic logic?  (I mean, there are two or three
different conventions for "and" and "or"!)

If we decide that "iff" is too confusing for people, the only benefit
to "<=>" is that it's practically guaranteed to be _more_ confusing
and nearly everyone will have to look it up!  Why don't we just
replace "iff" everywhere with "if", since people will get the idea?

After all, as fun as "iff" is to use, the original example wasn't
even completely clear there:

With a numeric argument, turn blinking cursor mode on iff ARG is positive.

Sure, blink-cursor-mode will change the blinking cursor to on if and
only iff ARG is positive, but it doesn't specify that it will turn the
blinking cursor off otherwise, since it's saying it will take a
certain action iff a condition is met.  It could just leave the cursor
in whatever state it's in if ARG is 0.

So, I think the whole argument about different representations of
"iff" is a waste of time.

-- 
Alan Shutko <ats@acm.org> - I am the rocks.
His ears made him look like a taxicab with both doors open

RE: typo in frame.el

[Drew Adams]

"Uncle!" (= I give up.)

Please don't take my Friday frivolity wrong. Work to make Emacs more
text-processing-oriented for non-programmers is wonderful. Emacs is great,
and will be better still with your additions, David.

And your last paragraph is very similar to what I wrote in the beginning
(see next), so I think we (at least the two of us) can agree on "iff". My
proposal of "<=>" was in the interest of "internationalism".

Here's what I wrote earlier:

> At the very worst, someone unacquainted with the meaning of "iff" will
> likely interpret it as "if" (and as an benign typo), so there is no harm
> done in using it: (the most important) half of its meaning is always
> conveyed.

Here is the need, as I see it: something short to convey "if and only if"
(equivalence). As I don't see a universally common candidate, I suggest that
whatever convention is used ("iff", whatever) is explained in the Emacs
manual. That's all.

And, if you use an abbreviation (like "iff") in the first doc-string line,
you can always explain things more clearly in the rest of the doc string
(e.g. "if and only if").

Cheers,

  - Drew

-----Original Message-----
From: dak@lola.goethe.zz [mailto:dak@lola.goethe.zz]On Behalf Of David
Kastrup
Sent: Friday, April 09, 2004 3:37 PM
To: Drew Adams
Cc: Simon Josefsson; emacs-devel@gnu.org
Subject: Re: typo in frame.el

...

Really, I don't know why I am working on getting AUCTeX and
preview-latex to create easy interfaces to WYSIWYG LaTeX creation.  I
am working hard to provide an increasingly compelling environment also
for non-programmers for text processing.

I just fail to see what purpose <=> would serve that could not be done
otherwise.  Even "iff" is wagonloads better.  People that don't
understand that terminology are probably prone to not interpreting
"if" in the mathematical way, anyhow.  So if they consider it as a
typo and apply the everyday meaning of "if" to it instead of the
mathematical meaning, they are pretty much fine.

Re: typo in frame.el

[David Kastrup]

"Drew Adams" <drew.adams@oracle.com> writes:

> Here's what I wrote earlier:
> 
> > At the very worst, someone unacquainted with the meaning of "iff"
> > will likely interpret it as "if" (and as an benign typo), so there
> > is no harm done in using it: (the most important) half of its
> > meaning is always conveyed.

That's different from what I wrote: I said that
non-matheticians/programmers/logicians use "if" in the meaning "iff"
more often than otherwise, so they get more than half of its meaning.

Anyway, in this case, the doc string is unclear anyway.

It should be

Turn cursor blink mode on if ARG is positive, off otherwise.

The problem is that the "iff" merely implies "Don't turn cursor blink
mode on if ARG is not positive" but that's different from turning it
off.  A programmer will probably make the right connection because it
is cool to do so, but a true mathematician will still get confused by
this wannabe mathematician jingo.

-- 
David Kastrup, Kriemhildstr. 15, 44793 Bochum

Re: typo in frame.el

[Alfred M. Szmidt]

   Whereas "iff" has been perfectly clear to me every time I've seen
   it.

I always get confused with it, since it could be a honest typo for a
plain "if".  Why not just use "if and only if", which is the meaning
of "iff"?

It is far clerer then "<=>" or any other funny ASCII combination, and
it also makes sense for anyone who is capable of reading some basic
English.

Cheers.

Re: typo in frame.el

[Miles Bader]

On Fri, Apr 09, 2004 at 01:46:53PM -0700, Drew Adams wrote:
> How about we use "<=>", conventionally, in doc strings, but we also explain
> this convention in the Emacs manual, for those who have never seen it?

No, I don't like it.  It's both ugly and confusing for the bulk of users
("explaining it in the manual" is not adequate).  It's also rather contrary
to the style used in doc-strings, which is rather resolutely textual, and
largely eschews any kind of pseudo-graphics.

I think there's nothing wrong with "iff".  As someone else pointed out, even
for those who don't know its exact meaning, it's a "benign" typo --
interpreting it as "if" is close enough for them to get the essential meaning
of the doc-string (something like "<=>" on the other hand, does not degrade
so gracefully).

-Miles
-- 
Run away!  Run away!

Re: typo in frame.el

[Juri Linkov]

David Kastrup <dak@gnu.org> writes:
> "Drew Adams" <drew.adams@oracle.com> writes:
>> Hmmm...  I suppose so. But two silly counter-arguments, then I'll rest:
>> 
>> 1) The unwashed masses seem to have digested ":)" OK...
>
> "Digested"?  I'd say they created it.  I doubt that hackers need
> explicit warning labels for nonobvious humor.

:-) is nothing more than a symbol to indicate an emotional state of
the sender.  And as in RL one can make jokes with a different face,
using smiley makes the same difference as for making jokes with
a smile or with a Good Friday face :-)

> I said that non-matheticians/programmers/logicians use "if" in the
> meaning "iff" more often than otherwise, so they get more than half
> of its meaning.
>
> Anyway, in this case, the doc string is unclear anyway.
>
> It should be
>
> Turn cursor blink mode on if ARG is positive, off otherwise.

Rephrasing it will undoubtedly help to reduce misunderstanding for
non-mathematicians/programmers/logicians, and it even don't have to
fit on one line in the given case, because it's not the first line of
a docstring.  But anyhow, it seems that eliminating "iff" is too late,
because "iff" is already too widespread in Emacs source files.

-- 
Juri Linkov
http://www.jurta.org/emacs/

Re: typo in frame.el

[Richard Stallman]

    I prefer "precisely when" instead of "if and only if".  Sometimes,
    even the "precisely" can be omitted.  Less technobabbly.  Alas,
    changing "iff" into "when" can make a documentation string line too
    long.

I think "precisely when" is a clear way to say it, but just "when"
often is not.