DIS: Re: BUS: Re: OFF: [Arbitor] Appeal of CFJ 3534; go vote!

[Josh T]

>  And finally, I learned about the difference between traditional
(tategaki)
and modern (yokogaki) ordering in Japanese today!  I'd always wondered
why
I was confused about that (looking at Japanese text from different
sources)
but never got around to looking it up.  So thanks for that :).

Your Arabic thing also solves my problem of trying to write a piece of text
in Japanese that makes sense when read in both directions the language can
be read.

天火狐

On 2 July 2017 at 15:12, Kerim Aydin  wrote:

>
>
> On Sun, 2 Jul 2017, omd wrote:
> > First of all, I'd like to note that Gmail displays the message
> > differently from ais523's images.  I see
> > {
> > [arabic text] : I call for judgement on the following statement
> > }
>
> Some evidence, and some commentary:
>
> First the gratuitous evidence for the record:
>
> I composed the message in a fixed-width font.  It appears for me, both in
> composition and in the message as received from the lists, the way it
> appears
> in the archives (using a reasonably-wide window), a single line with
>   [Latin] : [Arabic]
>
> https://mailman.agoranomic.org/cgi-bin/mailman/private/
> agora-business/2017-June/035191.html
>
> As "evidence", a note on my intent:  I had intended it to be
> center-justified,
> with the center as close to the colon as possible.  I thought about
> indenting it slightly on the left side to show that the intent to center
> it,
> but since the line was already beyond 80 characters, I wanted to minimize
> the
> chance across others' displays that it would wrap, so I didn't add the
> extra
> space. My editor and reader don't wrap until >120.
>
>
> Now some commentary (not official gratuitous arguments):
>
>   - We have a tradition of fixed-width displays, to the extent that we make
> our reports and other legal documents (tables) that way.  If tables
> don't
> display correctly, we tend to say "switch to a fixed width browser"
> not "we
> have to go by the quirks of individuals' mail clients".
>
>   - Not sure we've had a set standard on line length (differs by report),
> out
> of politeness definitely under 80.  That's the acknowledged weakness
> of my
> single line.
>
>   - Still, within the Rules we respect the authors' intent with ASCII art
> (and
> would frown at a rulekeepor who squeezed out the linespace in the Town
> Fountain as "inconsequential").  Of course, we've never used such
> positioning to make a legal distinction.
>
>   - In questions where it matters, it might be great to use this case to
> set
> a precedent.  The one I would suggest is "the way it displays in the
> archives is the canonical form" (without getting into whether the
> Distributor could mess with that one day :) ).
>
>   - And yes, I also acknowledge that narrowing the window when looking at
> the
> archives causes a line wrap - but the same is true for report tables.
> It's not an unreasonable hardship (IMO) to say "in doubt, view it in
> the
> archives with a window width sufficient to respect the author's fairly
> clear intent."
>
> That's all commentary on display, bytes, etc.  Now, assuming others are
> willing to judge it linguistically, as displayed on the archives with a
> sufficient window width (~90 characters are more):
>
>   - A main point for me is interpretation of the word "following".  I
> believe
> that a native Arabic-speaker would read "following" in the Arabic
> sentence as the thing past the colon (the Latin text), but I don't have
> a native-speaker on hand to ask.
>
>   - I think that CFJ 1267 and its two(!) appeals are the best discussion of
> using fixed-width "ASCII" art uncertainty to look at timing of actions.
> Definitely worth a look, including the controversy caused shown in the
> appeals:  https://faculty.washington.edu/kerim/nomic/cases/?1267
>
>   - Finally, since I kind of amped-up the emotion on this one, let me say I
> wouldn't be insulted if the whole thing was thrown out as ambiguous, or
> any particular interpretation (even favoring the Latin over the
> Arabic),
> as long as it's done on linguistic grounds and hopefully in a way that
> can apply to other/all languages similarly, or cover the mixing of
> multiple languages.
>
>   - And finally, I learned about the difference between traditional
> (tategaki)
> and modern (yokogaki) ordering in Japanese today!  I'd always wondered
> why
> I was confused about that (looking at Japanese text from different
> sources)
> but never got around to looking it up.  So thanks for that :).
>
>
>
>

DIS: Re: BUS: Re: OFF: [Arbitor] Appeal of CFJ 3534; go vote!

[Ørjan Johansen]

On Sun, 2 Jul 2017, omd wrote:


Oh, and I'd say Unicode basically doesn't matter, only visual
appearance in common email clients.  The following text:
{
‮‭I do X.‬ ‭I do Y.‬‬
}
uses the Unicode control characters LRO, RLO, and PDF; a pseudo-HTML
equivalent would be:
{
I do X. I do Y.
}
In other words, two LTR blocks within a RTL block.  Semantically, it's
quite clear that "I do X." comes first, but visually the text should
look identical to
{
I do Y. I do X.
}


FWIW, in my setup (alpine via tmux (optional, same without) and putty), i 
see X first.


Greetings,
Ørjan.

DIS: Re: BUS: Re: OFF: [Arbitor] Appeal of CFJ 3534; go vote!

[omd]

On Sun, Jul 2, 2017 at 9:12 PM, Kerim Aydin  wrote:
>   - We have a tradition of fixed-width displays, to the extent that we make
> our reports and other legal documents (tables) that way.  If tables don't
> display correctly, we tend to say "switch to a fixed width browser" not 
> "we
> have to go by the quirks of individuals' mail clients".

Just to be 100% clear: unlike ais523's example, the Gmail issue has
nothing to do with wrapping; it just displays in the wrong order.

IMO, there are two reasons why "look to the archives" is less
compelling in this case than in the case of fixed-width reports:

- With reports, you can see that they're mangled.  Here, the message
appears intact (just backwards from how it was meant to be displayed),
so there's no indication that checking the archives is necessary.

- Fixed-width reports are traditional, while Arabic is not; thus, we
should be be somewhat more deferential to the burden of handling
Arabic messages (including dealing with implementation issues) as
potentially unreasonable.

Re: DIS: Re: BUS: Re: OFF: [Arbitor] Appeal of CFJ 3534; go vote!

[omd]

On Sun, Jul 2, 2017 at 10:40 PM, Alex Smith  wrote:
> I'd say that your first two sentences are both effectively quoting the
> end of the sentence; «I disagree on this point, and the reason is
> "colons aren't always used for quotation"». The third is ambiguous
> between the meaning you intended, and what to me is a more reasonable
> meaning outside context, «H. Arbitor said "perhaps this discussion
> should be part of the record?"». It's also worth noting that it's a
> fairly tortured use of punctuation in all three sentences; the first
> would be more naturally written with a semicolon, the latter two with
> commas.

My impression is that even specifically in Agora, using colons to
address people is fairly common.  As for tortured: perhaps, but I do
use colons pretty often even when I'm not trying to make a point. :)
Indeed, too often.


> Imagine the following:
>
>> I do not do the following:
>> I call a CFJ on …
>
> That seems identical grammatically to your "Well then:" example, and
> yet I can't imagine it actually calling a CFJ.

I can see where you're coming from, but I disagree that "I do not do
the following" is grammatically similar to "Well then".  Going back to
your programming analogy, "I do not do the following" feels like it
*should* take a quote, or at least a lambda, whereas "Well then" is
more like a block, in the C sense.

And after all, in C-family languages, if we comment out the statement
(if, while, etc.) introducing a block, the result is to perform the
block once, unconditionally.  Since uninterpretable text in Agora is
effectively commented out, shouldn't we do the same? :)

Well, maybe that's taking the analogy a little too far.

Re: DIS: Re: BUS: Re: OFF: [Arbitor] Appeal of CFJ 3534; go vote!

[Alex Smith]

On Sun, 2017-07-02 at 22:30 +0200, omd wrote:
> > On Sun, Jul 2, 2017 at 7:25 PM, Alex Smith  wrote:
> > and in the first case, it's clear that the second line is
> > being quoted even if you don't attempt to machine-translate the Arabic
> > (from the fairly well-known facts that Arabic is right-to-left and that
> > a colon is a method of effectively quoting the rest of the sentence).
> 
> I disagree on this point: colons aren't always used for quotation.
> Therefore: given the inherent uncertainty in unknown text followed by
> a colon, I maintain that it should be treated as meaningless.  H.
> Arbitor: perhaps this discussion should be part of the record?

I'd say that your first two sentences are both effectively quoting the
end of the sentence; «I disagree on this point, and the reason is
"colons aren't always used for quotation"». The third is ambiguous
between the meaning you intended, and what to me is a more reasonable
meaning outside context, «H. Arbitor said "perhaps this discussion
should be part of the record?"». It's also worth noting that it's a
fairly tortured use of punctuation in all three sentences; the first
would be more naturally written with a semicolon, the latter two with
commas.

> > I'd be interested in what you make of the (currently unjudged) CFJ that
> > may or may not have been called in the following message by PSS:
> > […]
> > > Well then:
> > > 
> > > I invoke judgement on the other statement : I invoke judgement on the
> > > other statement
> 
> By the way, note that the "Well then:" is an example of a colon
> followed by an action.

It's an elision for "Well, then, I will do the following:". The
action's happening because the introduction implies that it is.

I guess perhaps "quote" is the wrong word. What the colon's doing is
causing the text before the colon to "operate" on the text after the
colon, perhaps modifying it or causing it to act in an unusual way. (In
a programming language, this would be represented by a block «{…}»
rather than a string literal «"…"».)

Imagine the following:

> I do not do the following:
> I call a CFJ on …

That seems identical grammatically to your "Well then:" example, and
yet I can't imagine it actually calling a CFJ.

-- 
ais523

Re: DIS: Re: BUS: Re: OFF: [Arbitor] Appeal of CFJ 3534; go vote!

[omd]

On Sun, Jul 2, 2017 at 7:25 PM, Alex Smith  wrote:
> and in the first case, it's clear that the second line is
> being quoted even if you don't attempt to machine-translate the Arabic
> (from the fairly well-known facts that Arabic is right-to-left and that
> a colon is a method of effectively quoting the rest of the sentence).

I disagree on this point: colons aren't always used for quotation.
Therefore: given the inherent uncertainty in unknown text followed by
a colon, I maintain that it should be treated as meaningless.  H.
Arbitor: perhaps this discussion should be part of the record?

> I'd be interested in what you make of the (currently unjudged) CFJ that
> may or may not have been called in the following message by PSS:
> […]
>> Well then:
>>
>> I invoke judgement on the other statement : I invoke judgement on the
>> other statement

By the way, note that the "Well then:" is an example of a colon
followed by an action.

> I'm having severe problems parsing this, and am leaning towards
> considering it meaningless. By your reasoning, though, this created two
> CFJs?

Possibly, but I think it's too ambiguous.

In my view,
{
I eat a pickle: I eat another pickle.
}
does not quote the second part - it's just two actions.  On the other hand,
{
I CFJ on the following statement: I eat another pickle.
}
obviously does quote the second part.  But what about this?
{
I CFJ on the other statement in this message: I eat another pickle.
}
The first part doesn't *need* a quote; without a colon, I would
interpret the second part as a standalone action that also happens to
be the target of an indirect reference from the first part:
{
I CFJ on the other statement in this message.  I eat another pickle.
}
But nor is it obvious that the first part can't take a quote, i.e.
that it can't be a wordier synonym of "the following statement".  The
colon weighs in favor of the quote interpretation, but the fact that
"I eat another pickle" is in the form of an action (well, if eating a
pickle were an Agoran action) weighs the other way.  Perhaps it is
ambiguous, or perhaps the colon wins out.  But in the actual message,
"the other statement" is vague enough to weigh in favor of
meaninglessness...

DIS: Re: BUS: Re: OFF: [Arbitor] Appeal of CFJ 3534; go vote!

[Alex Smith]

On Sun, 2017-07-02 at 18:21 +0200, omd wrote:
> Anyway, I'd argue that this doesn't matter, nor is it necessary to
> inspect the Unicode to determine the 'correct' ordering.  The effect
> of the message can be decided on much simpler grounds:
> 
> 1. The text "I call for judgement on the following statement" should
> be interpreted as an action, not implicitly quoted by the Arabic
> text.
> 
> This does not depend on the ordering; it would be true even if the
> Arabic version unambiguously came first, as in
> {
> :أدعو إلى إصدار حكم بشأن البيان التالي
> I call for judgement on the following statement
> }
> 
> [note: I don't believe that this would succeed in calling for
> judgement, as there would be no "following" statement, see below.]
> 
> Why?   Because we’d accept this as performing an action:
> {
> nkeplwgplxgioyzjvtxjnncsqscvntlbdqromyeyvlhkjgteaqnneqgujjpwcbyfrpueo
> ydjjk
> I do an action.
> }

Just to note that I explicitly disagree with this part of your
argument. We've long held that quotations of actions are not actually
performed, and in the first case, it's clear that the second line is
being quoted even if you don't attempt to machine-translate the Arabic
(from the fairly well-known facts that Arabic is right-to-left and that
a colon is a method of effectively quoting the rest of the sentence).

I'd consider "nkep: I do an action" to have no effect (and in
particular, not perform the action in question); to me, the most
sensible reading of that sentence is as a *definition* of nkep, and in
particular, the reading of the sentence as "«undefined word»(«quoted
action»)" is so strong that if it means anything at all, I'd expect it
to mean something of that form.

I'd be interested in what you make of the (currently unjudged) CFJ that
may or may not have been called in the following message by PSS:

On Wed, 2017-06-28 at 18:49 -0700, Publius Scribonius Scholasticus
wrote:
> Well then:
> 
> I invoke judgement on the other statement : I invoke judgement on the
> other statement

I'm having severe problems parsing this, and am leaning towards
considering it meaningless. By your reasoning, though, this created two
CFJs?

-- 
ais523