Mismodeling Indo-European Origins: The Assault on Historical Linguistics (video)

Arnaud Fournet

2017-04-24 12:41:35 UTC

Post by Yusuf B Gursey
http://youtu.be/hrQ_vgfkxNg

I'm interested in seeing which side Shithead PTD sides with.

A.

Arnaud Fournet

2017-04-24 12:46:12 UTC

Post by Yusuf B Gursey
http://youtu.be/hrQ_vgfkxNg

I'm really offended that they don't mention Hurrian and Urartian as members of PIE language family.
This is really nasty.
A.

DKleinecke

2017-04-24 17:19:46 UTC

Post by Arnaud Fournet

Post by Yusuf B Gursey
http://youtu.be/hrQ_vgfkxNg

I'm really offended that they don't mention Hurrian and Urartian as members of PIE language family.
This is really nasty.
A.

It looks like they never considered the possibility. The point
of view is quite conventional and conservative.

Useful YouTube to know about but I got bored halfway through
and stopped listening. People who don't know about historical
linguistics should find it interesting.

Peter T. Daniels

2017-04-24 20:05:50 UTC

Post by DKleinecke

Post by Arnaud Fournet

Post by Yusuf B Gursey
http://youtu.be/hrQ_vgfkxNg

I'm really offended that they don't mention Hurrian and Urartian as members of PIE language family.
This is really nasty.
A.

It looks like they never considered the possibility. The point
of view is quite conventional and conservative.
Useful YouTube to know about but I got bored halfway through
and stopped listening. People who don't know about historical
linguistics should find it interesting.

Sounds like it's ideal for Fournet, then. I don't do YouTubes.

Ruud Harmsen

2017-04-24 20:28:32 UTC

Mon, 24 Apr 2017 13:05:50 -0700 (PDT): "Peter T. Daniels"

Post by Peter T. Daniels
Sounds like it's ideal for Fournet, then. I don't do YouTubes.

And no Wikipedia and no nothing. An icon of the past. Missed
opportunities.

https://en.wikipedia.org/wiki/Chriet_Titulaer

--
Ruud Harmsen, http://rudhar.com

Peter T. Daniels

2017-04-24 20:45:16 UTC

Post by Ruud Harmsen
Mon, 24 Apr 2017 13:05:50 -0700 (PDT): "Peter T. Daniels"

Post by Peter T. Daniels
Sounds like it's ideal for Fournet, then. I don't do YouTubes.

And no Wikipedia and no nothing. An icon of the past. Missed
opportunities.
https://en.wikipedia.org/wiki/Chriet_Titulaer

And that's a reason for not following links. Of what earthly interest was that
gentleman to me or us, except that he lived until yesterday?

Ruud Harmsen

2017-04-25 05:13:37 UTC

Mon, 24 Apr 2017 13:45:16 -0700 (PDT): "Peter T. Daniels"

Post by Ruud Harmsen
Mon, 24 Apr 2017 13:05:50 -0700 (PDT): "Peter T. Daniels"

Post by Peter T. Daniels
Sounds like it's ideal for Fournet, then. I don't do YouTubes.

And no Wikipedia and no nothing. An icon of the past. Missed
opportunities.
https://en.wikipedia.org/wiki/Chriet_Titulaer

And that's a reason for not following links. Of what earthly interest was that
gentleman to me or us, except that he lived until yesterday?

So you did have a look.

--
Ruud Harmsen, http://rudhar.com

Peter T. Daniels

2017-04-25 13:21:34 UTC

Post by Ruud Harmsen
Mon, 24 Apr 2017 13:45:16 -0700 (PDT): "Peter T. Daniels"

Post by Ruud Harmsen
Mon, 24 Apr 2017 13:05:50 -0700 (PDT): "Peter T. Daniels"

Post by Peter T. Daniels
Sounds like it's ideal for Fournet, then. I don't do YouTubes.

And no Wikipedia and no nothing. An icon of the past. Missed
opportunities.
https://en.wikipedia.org/wiki/Chriet_Titulaer

And that's a reason for not following links. Of what earthly interest was that
gentleman to me or us, except that he lived until yesterday?

So you did have a look.

And I'll try to remember not to follow a link from you unless you say what
might be interesting about it.

Did you in fact have any reason at all for posting that link?

Wikiplinks don't cause computer problems, such as freezes or immense delays.

Ruud Harmsen

2017-04-25 17:41:00 UTC

Tue, 25 Apr 2017 06:21:34 -0700 (PDT): "Peter T. Daniels"

Post by Ruud Harmsen
Mon, 24 Apr 2017 13:45:16 -0700 (PDT): "Peter T. Daniels"

Post by Ruud Harmsen
Mon, 24 Apr 2017 13:05:50 -0700 (PDT): "Peter T. Daniels"

Post by Peter T. Daniels
Sounds like it's ideal for Fournet, then. I don't do YouTubes.

And no Wikipedia and no nothing. An icon of the past. Missed
opportunities.
https://en.wikipedia.org/wiki/Chriet_Titulaer

And that's a reason for not following links. Of what earthly interest was that
gentleman to me or us, except that he lived until yesterday?

So you did have a look.

And I'll try to remember not to follow a link from you unless you say what
might be interesting about it.
Did you in fact have any reason at all for posting that link?

No answer.

Post by Peter T. Daniels
Wikiplinks don't cause computer problems, such as freezes or immense delays.

They don't!? What does?

O wait, that was when your employer supplied you with a vastly
obsolete computer from the stone age, or they offered a better one but
you refused to use it. About 15 years ago.

--
Ruud Harmsen, http://rudhar.com

Peter T. Daniels

2017-04-25 19:12:23 UTC

Post by Ruud Harmsen
Tue, 25 Apr 2017 06:21:34 -0700 (PDT): "Peter T. Daniels"

Post by Ruud Harmsen
Mon, 24 Apr 2017 13:45:16 -0700 (PDT): "Peter T. Daniels"

Post by Ruud Harmsen
Mon, 24 Apr 2017 13:05:50 -0700 (PDT): "Peter T. Daniels"

Post by Peter T. Daniels
Sounds like it's ideal for Fournet, then. I don't do YouTubes.

And no Wikipedia and no nothing. An icon of the past. Missed
opportunities.
https://en.wikipedia.org/wiki/Chriet_Titulaer

And that's a reason for not following links. Of what earthly interest was that
gentleman to me or us, except that he lived until yesterday?

So you did have a look.

And I'll try to remember not to follow a link from you unless you say what
might be interesting about it.
Did you in fact have any reason at all for posting that link?

No answer.

You don't know why you posted it?

Post by Peter T. Daniels
Wikiplinks don't cause computer problems, such as freezes or immense delays.

They don't!? What does?
O wait, that was when your employer supplied you with a vastly
obsolete computer from the stone age, or they offered a better one but
you refused to use it. About 15 years ago.

Wrong, wrong, and wrong.

They forced me to switch from Mac to PC; they provided a used computer that was
on its last legs; I traded it in for a more recent one; I eventually got a laptop
that was inexpensive because it didn't have the latest up-to-date hardware; and
when I tried to replace the battery it turned out that the specified battery
model would not fit into the cradle: one of the vanes was a fraction of a mm
off, though the old battery slipped in and out without difficulty. So I put a
few more $ than the battery would have cost into a newer reconditioned model,
running Windows 8.1. When Windows 10 came along I took the free upgrade and
bought Office2013 -- unaware that Office 2016 would become available a little
over a month later. It then turned out that neither Windows 10 nor Word2013
can handle all the fonts I routinely use, so the only practical use the newer
laptop has is for taking to conferences where I can use it to give PowerPoints
without having to plug it in somewhere. And the first step in this sequence was
in 2006.

Not that any of that is any of your business.

Arnaud Fournet

2017-04-26 05:57:14 UTC

Post by Peter T. Daniels
Wrong, wrong, and wrong.
They forced me to switch from Mac to PC; they provided a used computer that was
on its last legs; I traded it in for a more recent one; I eventually got a laptop
that was inexpensive because it didn't have the latest up-to-date hardware; and
when I tried to replace the battery it turned out that the specified battery
model would not fit into the cradle: one of the vanes was a fraction of a mm
off, though the old battery slipped in and out without difficulty. So I put a
few more $ than the battery would have cost into a newer reconditioned model,
running Windows 8.1. When Windows 10 came along I took the free upgrade and
bought Office2013 -- unaware that Office 2016 would become available a little
over a month later. It then turned out that neither Windows 10 nor Word2013
can handle all the fonts I routinely use, so the only practical use the newer
laptop has is for taking to conferences where I can use it to give PowerPoints
without having to plug it in somewhere. And the first step in this sequence was
in 2006.
Not that any of that is any of your business.

It's nice that you are publicly confessing to being a complete idiot.
A.

Ruud Harmsen

2017-04-26 07:38:14 UTC

Post by Peter T. Daniels
Sounds like it's ideal for Fournet, then. I don't do YouTubes.

And no Wikipedia and no nothing. An icon of the past. Missed
opportunities.
https://en.wikipedia.org/wiki/Chriet_Titulaer

And that's a reason for not following links. Of what earthly interest was that
gentleman to me or us, except that he lived until yesterday?

So you did have a look.

And I'll try to remember not to follow a link from you unless you say what
might be interesting about it.
Did you in fact have any reason at all for posting that link?

No answer.

You don't know why you posted it?

I do, but I don't feel any need or urge to tell you. Because you could
easily find out yourself.

Post by Peter T. Daniels
Wikiplinks don't cause computer problems, such as freezes or immense delays.

They don't!? What does?
O wait, that was when your employer supplied you with a vastly
obsolete computer from the stone age, or they offered a better one but
you refused to use it. About 15 years ago.

Wrong, wrong, and wrong.

Nice. Three lefts make a right.

Post by Peter T. Daniels
They forced me to switch from Mac to PC; they provided a used computer that was
on its last legs; I traded it in for a more recent one; I eventually got a laptop
that was inexpensive because it didn't have the latest up-to-date hardware; and
when I tried to replace the battery it turned out that the specified battery
model would not fit into the cradle: one of the vanes was a fraction of a mm
off, though the old battery slipped in and out without difficulty. So I put a
few more $ than the battery would have cost into a newer reconditioned model,
running Windows 8.1. When Windows 10 came along I took the free upgrade and
bought Office2013 -- unaware that Office 2016 would become available a little
over a month later. It then turned out that neither Windows 10 nor Word2013
can handle all the fonts I routinely use, so the only practical use the newer
laptop has is for taking to conferences where I can use it to give PowerPoints
without having to plug it in somewhere. And the first step in this sequence was
in 2006.

Facts and details! Interesting.

Post by Peter T. Daniels
Not that any of that is any of your business.

Too late, now I know.

--
Ruud Harmsen, http://rudhar.com

Peter T. Daniels

2017-04-26 12:20:15 UTC

Post by Peter T. Daniels
Sounds like it's ideal for Fournet, then. I don't do YouTubes.

And no Wikipedia and no nothing. An icon of the past. Missed
opportunities.
https://en.wikipedia.org/wiki/Chriet_Titulaer

And that's a reason for not following links. Of what earthly interest was that
gentleman to me or us, except that he lived until yesterday?

So you did have a look.

And I'll try to remember not to follow a link from you unless you say what
might be interesting about it.
Did you in fact have any reason at all for posting that link?

No answer.

You don't know why you posted it?

I do, but I don't feel any need or urge to tell you. Because you could
easily find out yourself.

You know that I looked at the article and found nothing relating to language or linguistics.

Why did you post it?

Post by Peter T. Daniels
Wikiplinks don't cause computer problems, such as freezes or immense delays.

They don't!? What does?
O wait, that was when your employer supplied you with a vastly
obsolete computer from the stone age, or they offered a better one but
you refused to use it. About 15 years ago.

Wrong, wrong, and wrong.

Nice. Three lefts make a right.

Three wrongs don't make a right.

Facts and details! Interesting.

Post by Peter T. Daniels
Not that any of that is any of your business.

Too late, now I know.

So what?

Ruud Harmsen

2017-04-26 07:46:04 UTC

Tue, 25 Apr 2017 12:12:23 -0700 (PDT): "Peter T. Daniels"

Post by Peter T. Daniels
running Windows 8.1.

So did I! But that one wouldn't switch on anymore and I bought another
one that has Windows 10. Later on, I found that the failed computer
was still under warranty (2 years, not 1) and it was repaired (new
mother board) at no cost for me. But there is a fresh Windows 8.1 on
it, and it takes a lot of time to make it usable (7 and 10 are all
right, 8 is unusable).

Post by Peter T. Daniels
When Windows 10 came along I took the free upgrade and
bought Office2013 --

So did I!

Post by Peter T. Daniels
unaware that Office 2016 would become available a little
over a month later.

Never heard of that.

Post by Peter T. Daniels
It then turned out that neither Windows 10 nor Word2013
can handle all the fonts I routinely use,

Which ones, for example? My impression is that Windows (and so Word
and browsers) covers the complete Unicode range until and beyond
0xffff by now.

http://rudhar.com/lingtics/uniclnkl.htm
http://rudhar.com/cgi-bin/shunicod.cgi

Post by Peter T. Daniels
so the only practical use the newer
laptop has is for taking to conferences where I can use it to give PowerPoints
without having to plug it in somewhere. And the first step in this sequence was
in 2006.

--
Ruud Harmsen, http://rudhar.com

Peter T. Daniels

2017-04-26 12:25:28 UTC

Post by Ruud Harmsen
Tue, 25 Apr 2017 12:12:23 -0700 (PDT): "Peter T. Daniels"

Post by Peter T. Daniels
running Windows 8.1.

Post by Peter T. Daniels
When Windows 10 came along I took the free upgrade and
bought Office2013 --

So did I!

Post by Peter T. Daniels
unaware that Office 2016 would become available a little
over a month later.

Never heard of that.

Post by Peter T. Daniels
It then turned out that neither Windows 10 nor Word2013
can handle all the fonts I routinely use,

Which ones, for example? My impression is that Windows (and so Word
and browsers) covers the complete Unicode range until and beyond
0xffff by now.

I don't remember. I opened my book once and found that of the 100-odd fonts that I'd
copied over and installed, some didn't work.

I do know that Noto Sans Mandaic doesn't even work in Word2007 (even though in
a Preview window it does work) and Microsoft Himalayan -- the Tibetan font --
doesn't hold its combined shape when a file is closed and reopened (but reapplying
its character style makes it behave and allows a pdf of proper Tibetan to be created).

Post by Ruud Harmsen
http://rudhar.com/lingtics/uniclnkl.htm
http://rudhar.com/cgi-bin/shunicod.cgi

Again links without explanation.

Ruud Harmsen

2017-04-26 13:21:25 UTC

Wed, 26 Apr 2017 05:25:28 -0700 (PDT): "Peter T. Daniels"

Post by Ruud Harmsen
Tue, 25 Apr 2017 12:12:23 -0700 (PDT): "Peter T. Daniels"

Post by Peter T. Daniels
running Windows 8.1.

Post by Peter T. Daniels
When Windows 10 came along I took the free upgrade and
bought Office2013 --

So did I!

Post by Peter T. Daniels
unaware that Office 2016 would become available a little
over a month later.

Never heard of that.

Post by Peter T. Daniels
It then turned out that neither Windows 10 nor Word2013
can handle all the fonts I routinely use,

Which ones, for example? My impression is that Windows (and so Word
and browsers) covers the complete Unicode range until and beyond
0xffff by now.

I don't remember. I opened my book once and found that of the 100-odd fonts that I'd
copied over and installed, some didn't work.
I do know that Noto Sans Mandaic doesn't even work

You're right, 0840-085F Mandaic isn't supported by Windows 10.

Post by Peter T. Daniels
in Word2007

Fonts are in Windows, not Word, so that has nothing to do with it.

Post by Peter T. Daniels
(even though in
a Preview window it does work) and Microsoft Himalayan -- the Tibetan font --

http://unicode.org/charts/PDF/U0F00.pdf
That one IS supported.

Post by Peter T. Daniels
doesn't hold its combined shape when a file is closed and reopened (but reapplying
its character style makes it behave and allows a pdf of proper Tibetan to be created).

Post by Ruud Harmsen
http://rudhar.com/lingtics/uniclnkl.htm
http://rudhar.com/cgi-bin/shunicod.cgi

Again links without explanation.

I just used them to check Mandaic and Tibetan. Well, but if you don't
want to know and don't want to learn anything, that's your free
choice. Freedom is paramount.

Ignorance is bliss.

--
Ruud Harmsen, http://rudhar.com

Peter T. Daniels

2017-04-26 13:58:15 UTC

Post by Ruud Harmsen
Wed, 26 Apr 2017 05:25:28 -0700 (PDT): "Peter T. Daniels"

Post by Ruud Harmsen
Tue, 25 Apr 2017 12:12:23 -0700 (PDT): "Peter T. Daniels"

Post by Peter T. Daniels
running Windows 8.1.

Post by Peter T. Daniels
When Windows 10 came along I took the free upgrade and
bought Office2013 --

So did I!

Post by Peter T. Daniels
unaware that Office 2016 would become available a little
over a month later.

Never heard of that.

Post by Peter T. Daniels
It then turned out that neither Windows 10 nor Word2013
can handle all the fonts I routinely use,

Which ones, for example? My impression is that Windows (and so Word
and browsers) covers the complete Unicode range until and beyond
0xffff by now.

I don't remember. I opened my book once and found that of the 100-odd fonts that I'd
copied over and installed, some didn't work.
I do know that Noto Sans Mandaic doesn't even work

You're right, 0840-085F Mandaic isn't supported by Windows 10.

Post by Peter T. Daniels
in Word2007

Fonts are in Windows, not Word, so that has nothing to do with it.

You must be extrapolating. You are wrong: I have been told that the font works
in other word processors that I cannot afford to purchase.

Post by Peter T. Daniels
(even though in
a Preview window it does work) and Microsoft Himalayan -- the Tibetan font --

http://unicode.org/charts/PDF/U0F00.pdf
That one IS supported.

Post by Peter T. Daniels
doesn't hold its combined shape when a file is closed and reopened (but reapplying
its character style makes it behave and allows a pdf of proper Tibetan to be created).

Post by Ruud Harmsen
http://rudhar.com/lingtics/uniclnkl.htm
http://rudhar.com/cgi-bin/shunicod.cgi

Again links without explanation.

I just used them to check Mandaic and Tibetan. Well, but if you don't
want to know and don't want to learn anything, that's your free
choice. Freedom is paramount.
Ignorance is bliss.

Is it your claim that whatever you've cooked up on those web pages is better
than BabelMap?

If it told you that Microsoft Himalaya works properly in Word2007, it is wrong.

Ruud Harmsen

2017-04-26 19:17:40 UTC

Post by Ruud Harmsen
http://rudhar.com/lingtics/uniclnkl.htm
http://rudhar.com/cgi-bin/shunicod.cgi

Ignorance is bliss.

Is it your claim that whatever you've cooked up on those web pages is better
than BabelMap?

I don't know what BabelMap is, so your guess is as good as mine. Of
anyone's. Or whatever.

Post by Peter T. Daniels
If it told you that Microsoft Himalaya works properly in Word2007, it is wrong.

I don't what Microsoft Himalaya is. I do know that:
- when looking up Tibetan in my list of Unicode ranges at
http://rudhar.com/lingtics/uniclnkl.htm, I find that it's in
0F00-0FFF.

- when I try 0F00 as the starting point and ask to see 256 characters
in http://rudhar.com/cgi-bin/shunicod.cgi, I see the same character as
in the PDF I see by clicking through to
http://unicode.org/charts/PDF/U0F00.pdf from
http://rudhar.com/lingtics/uniclnkl.htm. So: Tibetan works in Windows
10.

- when I copy that page (including the characters) to Word 2013, they
work there too. (Font: Calibri)

- typing 0f00-left-alt-X in a new document, produces that first
Tibetan character. Font: Microsoft Himalaya.

- Copy paste also works in Notepad.

- Copy paste also works in Open Office 4.1, with the font Arial
Unicode MS

- Same in Libre Office 4.4

--
Ruud Harmsen, http://rudhar.com

Peter T. Daniels

2017-04-26 21:10:59 UTC

Post by Ruud Harmsen
http://rudhar.com/lingtics/uniclnkl.htm
http://rudhar.com/cgi-bin/shunicod.cgi

Ignorance is bliss.

Is it your claim that whatever you've cooked up on those web pages is better
than BabelMap?

I don't know what BabelMap is, so your guess is as good as mine. Of
anyone's. Or whatever.

lmgtfy:

http://www.babelstone.co.uk/Software/BabelMap.html

Post by Peter T. Daniels
If it told you that Microsoft Himalaya works properly in Word2007, it is wrong.

- when looking up Tibetan in my list of Unicode ranges at
http://rudhar.com/lingtics/uniclnkl.htm, I find that it's in
0F00-0FFF.
- when I try 0F00 as the starting point and ask to see 256 characters
in http://rudhar.com/cgi-bin/shunicod.cgi, I see the same character as
in the PDF I see by clicking through to
http://unicode.org/charts/PDF/U0F00.pdf from
http://rudhar.com/lingtics/uniclnkl.htm. So: Tibetan works in Windows
10.
- when I copy that page (including the characters) to Word 2013, they
work there too. (Font: Calibri)
- typing 0f00-left-alt-X in a new document, produces that first
Tibetan character. Font: Microsoft Himalaya.
- Copy paste also works in Notepad.
- Copy paste also works in Open Office 4.1, with the font Arial
Unicode MS
- Same in Libre Office 4.4

Did you type (or enter by clicking on their cells or by typingtheir Alt-X codes
or by copy-pasting) a Tibetan word? Such as, the standard example "bsgrubs", which
shows the maximum of two prescripts, one subscript, a vowel, and two postscripts?

Probably the characters all combined properly into the word.

But did you then close the Word2007 file and reopen it? Was it still a properly
composed word, or were all the glyphs strung out in a row?

It does me no good if Word2013 will hold the shape of Tibetan words if it won't
accept strings in whichever other fonts it wouldn't accept.

Typing a "Complex Script" language isn't just a matter of entering the characters. They
need to combine properly. The most familiar example is Arabic. But it's very, very far
from the only one.

Ruud Harmsen

2017-04-26 21:25:02 UTC

Wed, 26 Apr 2017 14:10:59 -0700 (PDT): "Peter T. Daniels"

Post by Peter T. Daniels
Typing a "Complex Script" language isn't just a matter of entering the characters. They
need to combine properly. The most familiar example is Arabic. But it's very, very far
from the only one.

Yes, Word use to get the order wrong.

Test: http://rudhar.com/sfreview/unigglen.htm in Firefox, copy paste
to Word 2013.

STILL COMPLETELY WRONG. Hebrew and Arabic. Wrong order of letters, at
the word level.

Typical Microsoft behaviour. Not doing the obvious, prefering
complexity over symplicity, thinking from programmers instead of from
users. Very disappointing, but it's what I expected.

--
Ruud Harmsen, http://rudhar.com

Ruud Harmsen

2017-04-26 21:31:02 UTC

Post by Ruud Harmsen
Test: http://rudhar.com/sfreview/unigglen.htm in Firefox, copy paste
to Word 2013.

Libre Office does a better job, but also does it wrong, in a different
way, notably with the Hebrew.
Open Office: same.

No, wait, I think they do it right.

--
Ruud Harmsen, http://rudhar.com

Peter T. Daniels

2017-04-26 21:51:09 UTC

Post by Ruud Harmsen
Test: http://rudhar.com/sfreview/unigglen.htm in Firefox, copy paste
to Word 2013.

Libre Office does a better job, but also does it wrong, in a different
way, notably with the Hebrew.
Open Office: same.
No, wait, I think they do it right.

I'm sorry you don't have the competence to handle Complex Script text.

Ruud Harmsen

2017-04-27 07:21:12 UTC

Post by Ruud Harmsen
Test: http://rudhar.com/sfreview/unigglen.htm in Firefox, copy paste
to Word 2013.

Libre Office does a better job, but also does it wrong, in a different
way, notably with the Hebrew.
Open Office: same.
No, wait, I think they do it right.

I'm sorry you don't have the competence to handle Complex Script text.

I just put entities in HTML, like α for Greek, &#x06nn; for
Arabic etc. That works fine in all know browsers, and has done so
since I wrote that test page http://rudhar.com/sfreview/unigglen.htm .
It is dated 13 October 2001, that's a whopping 15 years ago and more.

Also directly in UTF8: http://rudhar.com/sfreview/ungglutf.htm, which
has no table, just running text.

If copy&paste from that page to another Windows application put the
characters in the wrong order, that's bug. That Word 2013, after all
those years, still has that bug, I think is a shame.

Post by Peter T. Daniels
I'm sorry you don't have the competence to handle Complex Script text.

It has nothing to do with that, in sane Windows applications (there
many, some also by Microsoft) it works fine.

Tibetan probably also work well in Libre Office and Open Office, both
of which are free (and slow). I can't test that because I know nothing
about the script and the language.

--
Ruud Harmsen, http://rudhar.com

Ruud Harmsen

2017-04-27 07:30:05 UTC

Post by Ruud Harmsen
http://rudhar.com/sfreview/unigglen.htm .
http://rudhar.com/sfreview/ungglutf.htm

I must admit there is an error in the Hebrew text there: the full stop
is at the wrong end of the sentence. I probably should have use a
special right-to-left or Hebrew stop. Or indicate the text direction
in HTML, which can be done.

An Arabic full stop exists (06D4), but I don't see anything for Hebrew
here:
http://www.library.upenn.edu/sitedocs/tutorials/charentities/charentity-he-ar.html
.

--
Ruud Harmsen, http://rudhar.com

Peter T. Daniels

2017-04-27 11:10:12 UTC

Post by Ruud Harmsen
Test: http://rudhar.com/sfreview/unigglen.htm in Firefox, copy paste
to Word 2013.

Libre Office does a better job, but also does it wrong, in a different
way, notably with the Hebrew.
Open Office: same.
No, wait, I think they do it right.

I'm sorry you don't have the competence to handle Complex Script text.

My work is not done in "browsers." I write for publication.

Post by Ruud Harmsen
Also directly in UTF8: http://rudhar.com/sfreview/ungglutf.htm, which
has no table, just running text.
If copy&paste from that page to another Windows application put the
characters in the wrong order, that's bug. That Word 2013, after all
those years, still has that bug, I think is a shame.

GIGO.

Post by Peter T. Daniels
I'm sorry you don't have the competence to handle Complex Script text.

It has nothing to do with that, in sane Windows applications (there
many, some also by Microsoft) it works fine.
Tibetan probably also work well in Libre Office and Open Office, both
of which are free (and slow). I can't test that because I know nothing
about the script and the language.

I was once stuck with a Linux box for a few weeks. There are so many things standard
in Word that Open/LibreOffice couldn't do that it was useless.

Ruud Harmsen

2017-04-28 09:41:26 UTC

Thu, 27 Apr 2017 04:10:12 -0700 (PDT): "Peter T. Daniels"

Post by Ruud Harmsen
I just put entities in HTML, like α for Greek, &#x06nn; for
Arabic etc. That works fine in all know browsers, and has done so
since I wrote that test page http://rudhar.com/sfreview/unigglen.htm .
It is dated 13 October 2001, that's a whopping 15 years ago and more.

My work is not done in "browsers." I write for publication.

So do I. In HTML. Paper publication is obsolete. An intermediate form
(although with disadvantages) is PDF, but that can be created from
HTML too.

GIGO.

Garbage in, garbage out.

How do you know where exactly the garbage is?

In my HTML? Entities are officially accepted in standards by W3.org.
So is UTF8.

In the text rendered by Firefox? The results look like good readible
Arabic/Hebrew, so no junk in sight there.

In the Paste Buffer after Ctrl-V (or Ctrl-Insert)? Do you have a tool
to see exactly what's in there?

In Word 2013 after the Paste (Ctrl-C or Shift-Insert). Clearly yes,
because the result is in the wrong order and with unconnected wrongly
formed Arabic characters.

So where is the error and what could be the fix?

Post by Peter T. Daniels
I was once stuck with a Linux box for a few weeks. There are so many things standard
in Word that Open/LibreOffice couldn't do that it was useless.

It doesn't do macros, for example. In a former situation I needed that
for translation work, because one client relied on it. Meanwhile they
no longer do.

--
Ruud Harmsen, http://rudhar.com

Peter T. Daniels

2017-04-26 21:50:35 UTC

Post by Ruud Harmsen
Wed, 26 Apr 2017 14:10:59 -0700 (PDT): "Peter T. Daniels"

Yes, Word use to get the order wrong.
Test: http://rudhar.com/sfreview/unigglen.htm in Firefox, copy paste
to Word 2013.
STILL COMPLETELY WRONG. Hebrew and Arabic. Wrong order of letters, at
the word level.
Typical Microsoft behaviour. Not doing the obvious, prefering
complexity over symplicity, thinking from programmers instead of from
users. Very disappointing, but it's what I expected.

I've never had a moment's trouble typing Hebrew, Arabic, or Syriac
in Word2007, and only
a moment's trouble with any of the Indic scripts (because you need to locate the virama
key, which isn't consistently in the same place for every script).

You may have copied faultily -- missing out the "R-to-L" code at the beginning and
the "L-to-R" code at the end.

Ruud Harmsen

2017-04-27 07:35:04 UTC

Wed, 26 Apr 2017 14:50:35 -0700 (PDT): "Peter T. Daniels"

Post by Peter T. Daniels
You may have copied faultily -- missing out the "R-to-L" code at the beginning and
the "L-to-R" code at the end.

In my HTML, there is no such code. It isn't necessary, because the
code ranges already make clear that the text is intended as
right-to-left, although it is encoded in HTML in the normal sequence.

Setting the control language to Hebrew, after pasting in Word, doesn't
fix the direction. Next Í'll try some copy pastes from Wikipedia.

--
Ruud Harmsen, http://rudhar.com

Ruud Harmsen

2017-04-27 07:46:39 UTC

Post by Ruud Harmsen
Setting the control language to Hebrew, after pasting in Word, doesn't
fix the direction. Next Í'll try some copy pastes from Wikipedia.

https://en.wikipedia.org/wiki/Tel_Aviv
Firefox to Word 2013: Same problem, Arabic and Hebrew for Tel Aviv are
in the wrong order.

All Hebrew page
https://he.wikipedia.org/wiki/%D7%AA%D7%9C_%D7%90%D7%91%D7%99%D7%91-%D7%99%D7%A4%D7%95
Same thing: wrong order.

So it is still my opinion: Word is a Stupid Program.

Peter T. Daniels

2017-04-27 11:13:09 UTC

Post by Ruud Harmsen
Setting the control language to Hebrew, after pasting in Word, doesn't
fix the direction. Next Í'll try some copy pastes from Wikipedia.

https://en.wikipedia.org/wiki/Tel_Aviv
Firefox to Word 2013: Same problem, Arabic and Hebrew for Tel Aviv are
in the wrong order.
All Hebrew page
https://he.wikipedia.org/wiki/%D7%AA%D7%9C_%D7%90%D7%91%D7%99%D7%91-%D7%99%D7%A4%D7%95
Same thing: wrong order.
So it is still my opinion: Word is a Stupid Program.

Firefox is incompetent and that makes Word stupid??

Word works perfectly for creating text in Complex Script languages. That you
don't know how to use the tool is not the tool's fault.

Ruud Harmsen

2017-04-28 09:33:45 UTC

Thu, 27 Apr 2017 04:13:09 -0700 (PDT): "Peter T. Daniels"

Post by Ruud Harmsen
So it is still my opinion: Word is a Stupid Program.

Firefox is incompetent and that makes Word stupid??

What makes you think that the problem is in Firefox? Did you do a
software analysis, looking at intermediate results, to get to know
more?
(See also Tak To further on the thread, who did.)

Post by Peter T. Daniels
Word works perfectly for creating text in Complex Script languages. That you
don't know how to use the tool is not the tool's fault.

An experiment I just did: seeing my encoding of the Arabic in
http://rudhar.com/sfreview/unigglen.htm, which is
فصيح
I entered manually in Word 2013 (not having a keyboard lay-out for
Arabic nor the wish to install one, although that is quite feasible in
Windows), the followed keystrokes:

0641 Alt-X 0635 Alt-X 064A Alt-X 062D Alt-X.

Result: the Arabic word faSiiH, with the correctly formed characters
in the correct order and connected as they should.

There was no need to specify the language first, nor to specify that I
was going to enter right-to-left (RTL) text. So apparently, in this
case Word does what in my view it should do (and what is also what
Firefox does when rendering HTML): understand that if someone enters
characters that are in any of the Unicode range for the Arabic script,
RTL is needed, simply because any language that uses the Arabic script
is always written RTL.

That this doesn't happen when that same text comes from the paste
buffer, I still think is a bug in Word. Default behaviour should
always lead to correct results as much as possible. Default behaviour
that predictably and testibly leads to incorrect results, is not a
good default and not a good software design choice.

--
Ruud Harmsen, http://rudhar.com

Peter T. Daniels

2017-04-28 12:21:44 UTC

Post by Ruud Harmsen
Thu, 27 Apr 2017 04:13:09 -0700 (PDT): "Peter T. Daniels"

Post by Ruud Harmsen
So it is still my opinion: Word is a Stupid Program.

Firefox is incompetent and that makes Word stupid??

What makes you think that the problem is in Firefox? Did you do a
software analysis, looking at intermediate results, to get to know
more?
(See also Tak To further on the thread, who did.)

Tak To seems to be pointing out deficiencies in Firefox.

Post by Peter T. Daniels
Word works perfectly for creating text in Complex Script languages. That you
don't know how to use the tool is not the tool's fault.

An experiment I just did: seeing my encoding of the Arabic in
http://rudhar.com/sfreview/unigglen.htm, which is
فصيح
I entered manually in Word 2013 (not having a keyboard lay-out for
Arabic nor the wish to install one, although that is quite feasible in
0641 Alt-X 0635 Alt-X 064A Alt-X 062D Alt-X.
Result: the Arabic word faSiiH, with the correctly formed characters
in the correct order and connected as they should.
There was no need to specify the language first, nor to specify that I
was going to enter right-to-left (RTL) text. So apparently, in this
case Word does what in my view it should do (and what is also what
Firefox does when rendering HTML): understand that if someone enters
characters that are in any of the Unicode range for the Arabic script,
RTL is needed, simply because any language that uses the Arabic script
is always written RTL.

You don't need to laboriously type Alt-X codes. You can simply open Insert Symbol
to the appropriate Unicode range, displaying the font you prefer, and click on the letters.

Now that you've finally used MSWord the way it's meant to be used, you've learned that
it handles the more usual Complex Scripts just fine, and your ire is misplaced. If
Firefox also accepts Complex Script input and flawlessly produces text in the language
in question, then clearly it's User Error if you're unable to Copy-Paste from one into
the other.

Post by Ruud Harmsen
That this doesn't happen when that same text comes from the paste
buffer, I still think is a bug in Word. Default behaviour should
always lead to correct results as much as possible. Default behaviour
that predictably and testibly leads to incorrect results, is not a
good default and not a good software design choice.

Why would what you do be "default behavior"?

Ruud Harmsen

2017-04-28 12:39:46 UTC

Fri, 28 Apr 2017 05:21:44 -0700 (PDT): "Peter T. Daniels"

Post by Ruud Harmsen
(See also Tak To further on the thread, who did.)

Tak To seems to be pointing out deficiencies in Firefox.

That is far from certain. Read again.

--
Ruud Harmsen, http://rudhar.com

Ruud Harmsen

2017-04-28 12:48:32 UTC

Fri, 28 Apr 2017 05:21:44 -0700 (PDT): "Peter T. Daniels"

Why would what you do be "default behavior"?

I am referring to SOFTWARE behaviour, of course, not MY behaviour.

--
Ruud Harmsen, http://rudhar.com

Ruud Harmsen

2017-04-28 12:50:51 UTC

Post by Peter T. Daniels
Why would what you do be "default behavior"?

I am referring to SOFTWARE behaviour, of course, not MY behaviour.

https://www.letras.mus.br/the-beatles/86319/

Sigh, it is indeed "junior". All my life (since 1964, probably), I was
convinced that those lyrics went: "now, Julia, behave yourself."!

--
Ruud Harmsen, http://rudhar.com

Ruud Harmsen

2017-04-28 12:52:13 UTC

Post by Peter T. Daniels
Why would what you do be "default behavior"?

I am referring to SOFTWARE behaviour, of course, not MY behaviour.

https://www.letras.mus.br/the-beatles/86319/
http://youtu.be/qPhPLH8wnro
Sigh, it is indeed "junior". All my life (since 1964, probably), I was
convinced that those lyrics went: "now, Julia, behave yourself."!

I'm not the only one:
http://www.amiright.com/misheard/song/badboy.shtml

--
Ruud Harmsen, http://rudhar.com

Ruud Harmsen

2017-04-28 13:05:23 UTC

Fri, 28 Apr 2017 05:21:44 -0700 (PDT): "Peter T. Daniels"

Post by Ruud Harmsen
There was no need to specify the language first, nor to specify that I
was going to enter right-to-left (RTL) text. So apparently, in this
case Word does what in my view it should do (and what is also what
Firefox does when rendering HTML): understand that if someone enters
characters that are in any of the Unicode range for the Arabic script,
RTL is needed, simply because any language that uses the Arabic script
is always written RTL.

You don't need to laboriously type Alt-X codes. You can simply open Insert Symbol
to the appropriate Unicode range, displaying the font you prefer, and click on the letters.

That's also a method, yes, I know about that. But it's almost as
tedious, especially in cases where I had already looked up the Unicode
scalars for use in a web page.

Post by Peter T. Daniels
Now that you've finally used MSWord the way it's meant to be used,

That has nothing to do with it.

Are you aware that you are never looking for knowledge and solutions,
but ONLY and ALWAYS for opportunities to humiliate other people? Why
do you try to do that? (I'm immune to it anyway, but still). Is it
some defect in your character? Or a trauma maybe?

Post by Peter T. Daniels
you've learned that
it handles the more usual Complex Scripts just fine, and your ire is misplaced.

I solemnly declare that the buggy behaviour was actually there a few
days ago, when I mentioned it, and it was there several times before,
over the years. I have seen it with my own eyes.

And no it's gone. I don't know why. Erratic bugs are the hardest to
fix. I know that from experience as a software engineer.

Post by Peter T. Daniels
If
Firefox also accepts Complex Script input

A browser is primarily for displaying text, not for inputting text.
Although it can. Test:
http://rudhar.com/sfreview/unigglen.htm =>
http://rudhar.com/cgi-bin/umlaut.cgi
Works

http://rudhar.com/sfreview/unigglen.htm =>
Word 2013 =>
http://rudhar.com/cgi-bin/umlaut.cgi
Works

Post by Peter T. Daniels
and flawlessly produces text in the language
in question, then clearly it's User Error if you're unable to Copy-Paste from one into
the other.

By the way, YOU were the one who complained that Word can't do certain
languages and fonts, remember? You never provided adequate data to
make the problem reproducible. I did, my problem mysteriously
vanished, and now _I_ am stupid. Right.

--
Ruud Harmsen, http://rudhar.com

Peter T. Daniels

2017-04-28 13:47:23 UTC

Post by Ruud Harmsen
By the way, YOU were the one who complained that Word can't do certain
languages and fonts, remember? You never provided adequate data to
make the problem reproducible.

? I told you the names of the two fonts that misbehave. I have found no other
font for either Mandaic or Tibetan to discover where the problem might lie.

What further data would be needed for "adequacy"?

I could also mention Avestan, but the Unicode-range font was done at my request
by a hobbyist who had long ago made a pre-Unicode font (her glyphs are accurate,
unlike some of those in Unicode's display), and she may not have been aware of
some subtlety required in the coding. Avestan is like Hebrew: right-to-left,
but no combining forms.

Post by Ruud Harmsen
I did, my problem mysteriously
vanished, and now _I_ am stupid. Right.

No idea what your ptoblem may have been.

I've occasionally copy-pasted Qur'an verses from web sites and never had a fant problem.

Ruud Harmsen

2017-04-28 18:36:00 UTC

Fri, 28 Apr 2017 06:47:23 -0700 (PDT): "Peter T. Daniels"

Avestan is like Hebrew: right-to-left, but no combining forms.

Like Hebrew??

http://unicode.org/charts/PDF/U10B00.pdf

--
Ruud Harmsen, http://rudhar.com

Peter T. Daniels

2017-04-28 21:27:06 UTC

Post by Ruud Harmsen
Fri, 28 Apr 2017 06:47:23 -0700 (PDT): "Peter T. Daniels"

Avestan is like Hebrew: right-to-left, but no combining forms.

Like Hebrew??
http://unicode.org/charts/PDF/U10B00.pdf

In what way, other than the shapes of the glyphs, does the Avestan alphabet differ
from the Hebrew?

Ruud Harmsen

2017-04-29 05:30:14 UTC

Fri, 28 Apr 2017 14:27:06 -0700 (PDT): "Peter T. Daniels"

Post by Ruud Harmsen
Fri, 28 Apr 2017 06:47:23 -0700 (PDT): "Peter T. Daniels"

Avestan is like Hebrew: right-to-left, but no combining forms.

Like Hebrew??
http://unicode.org/charts/PDF/U10B00.pdf

In what way, other than the shapes of the glyphs, does the Avestan alphabet differ
from the Hebrew?

I read:
AVESTAN LETTER AAO
AVESTAN LETTER AN
AVESTAN LETTER AAN

Does Hebrew have such things?

AVESTAN LETTER XYE
AVESTAN LETTER XVE
AVESTAN LETTER TTE

AVESTAN LETTER NGE
AVESTAN LETTER NGYE
AVESTAN LETTER NGVE
AVESTAN LETTER NE
AVESTAN LETTER NYE

Does Hebrew have such things?

Perhaps you don't follow the link and you are talking about a
completely different Avestan than I am?

--
Ruud Harmsen, http://rudhar.com

Peter T. Daniels

2017-04-29 12:56:33 UTC

Post by Ruud Harmsen
Fri, 28 Apr 2017 14:27:06 -0700 (PDT): "Peter T. Daniels"

Post by Ruud Harmsen
Fri, 28 Apr 2017 06:47:23 -0700 (PDT): "Peter T. Daniels"

Avestan is like Hebrew: right-to-left, but no combining forms.

Like Hebrew??
http://unicode.org/charts/PDF/U10B00.pdf

In what way, other than the shapes of the glyphs, does the Avestan alphabet differ
from the Hebrew?

AVESTAN LETTER AAO
AVESTAN LETTER AN
AVESTAN LETTER AAN
Does Hebrew have such things?
AVESTAN LETTER XYE
AVESTAN LETTER XVE
AVESTAN LETTER TTE
AVESTAN LETTER NGE
AVESTAN LETTER NGYE
AVESTAN LETTER NGVE
AVESTAN LETTER NE
AVESTAN LETTER NYE
Does Hebrew have such things?
Perhaps you don't follow the link and you are talking about a
completely different Avestan than I am?

As I asked, what is the difference -- in computerization terms, which is all
we've been talking about -- between Avestan and Hebrew? They are both scripts
that go right to left and have no combining forms.

Computers neither know nor care that Avestan is an alphabet and Hebrew is an
abjad, or that Avestan has more than twice as many letters as Hebrew.

In fact, though, graphonimically speaking, they seem uniquely to share the
property that (when the Hebrew is vocalized) they denote subphonemic distinctions
-- that in fact we are unable to identify satisfactorily.

Tak To

2017-04-27 02:02:00 UTC

Post by Ruud Harmsen
Wed, 26 Apr 2017 14:10:59 -0700 (PDT): "Peter T. Daniels"

I assume you meant "right letter order but wrong word order", because
that is what I see in Word 2007.

In any case, given that the source HTML has neither a "dir" attribute
or a "direction" property[1], what makes you think that Firebox is
behaving correctly (and MS Word not) in the first place?

Incidentally, if one copies the entire table and not just the Hebrew
or Arabic text in the cell, the word order would be correct.

[1] ignoring the more complicated "bidirectional text" schemes for
the time being...

Post by Ruud Harmsen
Typical Microsoft behaviour. Not doing the obvious, prefering
complexity over symplicity, thinking from programmers instead of from
users. Very disappointing, but it's what I expected.

It is not clear why you think it is "typical Microsoft" or "thinking
from [the point of view of] programmers". I don't care about the
first but being a programmer, I resent my brethren being blamed
for things that are not their faults.

I am sure you are aware that different "users" might have very
different ideas of what "simple" means.

--
Tak
----------------------------------------------------------------+-----
Tak To ***@alum.mit.eduxx
--------------------------------------------------------------------^^
[taode takto ~{LU5B~}] NB: trim the xx to get my real email addr

Ruud Harmsen

2017-04-27 07:37:50 UTC

Good question. In HTML I put the entities or UTF-8 character in the
normal order, that is: first characters first. The convention I think
is that direction is then switched when rendering. The behaviour of
Notepad (also by Microsoft) and Open Office / Libre Office support
that assumption.

--
Ruud Harmsen, http://rudhar.com

Tak To

2017-04-27 18:55:28 UTC

Good question. In HTML I put the entities or UTF-8 character in the
normal order, that is: first characters first.

Yes, I can see that by looking at the HTML source from Firefox.
However, note that this is not necessarily how Firefox stores
them internally, or what actually gets sent to the clipboard.
(more below)

Post by Ruud Harmsen
The convention I think
is that direction is then switched when rendering. The behaviour of
Notepad (also by Microsoft) and Open Office / Libre Office support
that assumption.

The Windows Cut&Paste (Clipboard) mechanism support formats other
than just (unicode) character stream. The receiving program
queries what formats are available for the current entity in
the clipboard, and then tells the sending program which format
is to be used. Between Firefox and MS Word two formats are
possible: "HTML Format" or "Unformatted Unicode Text". The
default is "HTML Format". (You can check this by doing a "Paste
Special".) OTOH, Notepad does not take "HTML Format" so there
lies difference. I have no idea what Libre Office does.

I believe the following is happening:

(A) When Firefox is told to send in "Unformatted Unicode Text"
format, it sends them out in "rendered order" rather than
"source order". Since there is no text direction (i.e., word
order) property in the original HTML code, Firefox has to use
heuristics, and somehow it decides to apply RTL order to the
entire string between the first Hebrew character and the last
one, including all the spaces in between while leaving the last
period (full-stop, ".") alone.

(B) When Firefox is told to send in "HTML Code" format, it sends
them out in "source order", retaining as many HTML elements
as possible.

(B1) When copying a chunk of Hebrew text in HTML, MS Word is
given only the inline HTML elements and none of the surrounding
blocks[1]. Lacking any text direction property in the inline
elements, MS Word decides to apply RTL to the individual
chunks of contiguous Hebrew characters, thereby retaining the
word order in the source.

(B2) When copying the entire table, MS is given additional
information about the table cell structures (<td>s),
in particular that cell containing the Hebrew characters
has a "align-right" attribute. It is possible that with
this additional information, MS decides to use a different
strategy and ends up with the same the order as in (A).

[1] It is not clear how much CSS information is sent from
Firefox. In any case HTML is organized in hierarchical
structures, so the precise semantics of a detached inner
structure is not well defined.

I hope you realize that the strategy employed in (B1) is
in fact much "simpler" than that in (A) or (B2). The latter
two, for example, require the knowledge that spaces and periods
are to be handled differently. This leads to questions
about other characters such as "$" or "*", etc.

Ruud Harmsen

2017-04-28 09:46:28 UTC

Post by Tak To
The Windows Cut&Paste (Clipboard) mechanism support formats other
than just (unicode) character stream. The receiving program
queries what formats are available for the current entity in
the clipboard, and then tells the sending program which format
is to be used.

I don't think that's possible, because when I copy (i.e. send text to
the clipboard), it is not yet know in which other program (or rather:
programs, because multiple pastes are possible) I will be doing the
paste or pastes. So the programs can't communicate because they don't
know what to communicate with.

I can even copy in Firefox, close Firefox, then open Word, and paste
in Word. Word could not talk to Firefox, because Firefox isn't active
at the time of the paste.

--
Ruud Harmsen, http://rudhar.com

Tak To

2017-04-28 17:36:36 UTC

I don't think that's possible, because when I copy (i.e. send text to
programs, because multiple pastes are possible) I will be doing the
paste or pastes. So the programs can't communicate because they don't
know what to communicate with.
I can even copy in Firefox, close Firefox, then open Word, and paste
in Word. Word could not talk to Firefox, because Firefox isn't active
at the time of the paste.

I might have given the false impression that the receiving
program queries the sending program directly, but the fact
is that the receiving program queries the clipboard.

That said, nowadays Windows programs comprise a large number
of dynamically linked library modules (DLLs) in addition to
the main executable (EXE). It is possible, and indeed rather
common for program A to call the program B's DLLs without
first starting B.EXE.

Note that what get put into the Clipboard are "objects" and
not just passive data. Each object type is registered with
the Clipboard so that the Clipboard knows which DLL to run
when an object is being "pasted". An object may share data
with another object, or may store data in temporary files.
Thus, putting multiple (types of) objects into the clipboard
needs not involve doing vast amount of memory transfer.

In all likelihood both Firefox and MS Word use standard
Windows DLLs to handle the common Clipboard object types
such as "Unformatted Unicode Text".

Ruud Harmsen

2017-04-28 18:48:02 UTC

Post by Ruud Harmsen
I can even copy in Firefox, close Firefox, then open Word, and paste
in Word. Word could not talk to Firefox, because Firefox isn't active
at the time of the paste.

I might have given the false impression that the receiving
program queries the sending program directly, but the fact
is that the receiving program queries the clipboard.
That said, nowadays Windows programs comprise a large number
of dynamically linked library modules (DLLs) in addition to
the main executable (EXE). It is possible, and indeed rather
common for program A to call the program B's DLLs without
first starting B.EXE.
Note that what get put into the Clipboard are "objects" and
not just passive data. Each object type is registered with
the Clipboard so that the Clipboard knows which DLL to run
when an object is being "pasted". An object may share data
with another object, or may store data in temporary files.
Thus, putting multiple (types of) objects into the clipboard
needs not involve doing vast amount of memory transfer.
In all likelihood both Firefox and MS Word use standard
Windows DLLs to handle the common Clipboard object types
such as "Unformatted Unicode Text".

Yes, that seems plausible.

--
Ruud Harmsen, http://rudhar.com

Ruud Harmsen

2017-04-28 09:48:17 UTC

Post by Tak To
Between Firefox and MS Word two formats are
possible: "HTML Format" or "Unformatted Unicode Text". The
default is "HTML Format". (You can check this by doing a "Paste
Special".) OTOH, Notepad does not take "HTML Format" so there
lies difference. I have no idea what Libre Office does.

In both cases, the pieces of Arabic text are identifiable as Arabic
text, so should be pasted, or rendered, or both, as Arabic text, which
is always right-to-left.

--
Ruud Harmsen, http://rudhar.com

Ruud Harmsen

2017-04-28 09:54:31 UTC

Post by Tak To
(B1) When copying a chunk of Hebrew text in HTML, MS Word is
given only the inline HTML elements and none of the surrounding
blocks[1]. Lacking any text direction property in the inline
elements, MS Word decides to apply RTL to the individual
chunks of contiguous Hebrew characters, thereby retaining the
word order in the source.

Good theory. Except that Word doesn't.

(Testing again.)
But it does! You're right!

But: it is now ALSO correct when I copy several cells from the
webpage! AND when I copy the whole page!

So they (Mozilla? Microsoft?) fixed the bug overnight and secretly
updated my system. I'm stumped.

--
Ruud Harmsen, http://rudhar.com

Ruud Harmsen

2017-04-28 09:55:56 UTC

Post by Tak To
(B2) When copying the entire table, MS is given additional
information about the table cell structures (<td>s),
in particular that cell containing the Hebrew characters
has a "align-right" attribute. It is possible that with
this additional information, MS decides to use a different
strategy and ends up with the same the order as in (A).

Possibly. But I didn't enter any such align-right or right-to-left
attributes anywhere. They are implicit, because the characters are
Hebrew.

--
Ruud Harmsen, http://rudhar.com

Tak To

2017-04-28 17:51:00 UTC

Possibly. But I didn't enter any such align-right or right-to-left
attributes anywhere. They are implicit, because the characters are
Hebrew.

I am not sure how you created your webpage, but in the HTML
source, the table cell containing the Hebrew text is assigned
to be of CSS class "MiddleEast"

<td class="MiddleEast">

And at the beginning of the HTML file the MiddleEast class is
define as

<style type="text/css">
.MiddleEast {
text-align: right;
font-size: 24pt; width=150;
vertical-align: top;
}
...
</style>

Hope this helps.

Ruud Harmsen

2017-04-28 18:53:59 UTC

Post by Ruud Harmsen
Possibly. But I didn't enter any such align-right or right-to-left
attributes anywhere. They are implicit, because the characters are
Hebrew.

I am not sure how you created your webpage,

By simple text editor, directly writing HTML and CSS.

http://rudhar.com/sfreview/html_en/

Post by Tak To
but in the HTML
source, the table cell containing the Hebrew text is assigned
to be of CSS class "MiddleEast"
<td class="MiddleEast">
And at the beginning of the HTML file the MiddleEast class is
define as
<style type="text/css">
.MiddleEast {
text-align: right;
font-size: 24pt; width=150;
vertical-align: top;
}
...
</style>

Yes, you're right, I forgot I ever did that (over 15 years ago), but I
did. Or rather, I wouldn't think I already did that then, I thought it
was much more recent.

I seem to be getting old.

--
Ruud Harmsen, http://rudhar.com

Ruud Harmsen

2017-04-28 09:58:42 UTC

Post by Tak To
I hope you realize that the strategy employed in (B1) is
in fact much "simpler" than that in (A) or (B2). The latter
two, for example, require the knowledge that spaces and periods
are to be handled differently. This leads to questions
about other characters such as "$" or "*", etc.

True.

To make it more complicated, Arabic has its own special comma
character, shaped as a 9 mirrored against a vertical axis; while
modern Hebrew uses the normal western script comma, which point "the
wrong way" (in my eyes, but as this is normal practice, it is in fact
OK).

--
Ruud Harmsen, http://rudhar.com

Tak To

2017-04-28 16:58:04 UTC

True.
To make it more complicated, Arabic has its own special comma
character, shaped as a 9 mirrored against a vertical axis; while
modern Hebrew uses the normal western script comma, which point "the
wrong way" (in my eyes, but as this is normal practice, it is in fact
OK).

In CJK we have a set of "full-width" punctuation marks in
addition to the "half-width" or Latin ones, and some
programs keep converting all full-width punctuation marks
to their half-width counterparts, thereby messing up the
alignment of characters.

There is also a set of vertical punctuation marks for
vertically running text...

Dingbat

2017-05-16 17:31:10 UTC

True.
To make it more complicated, Arabic has its own special comma
character, shaped as a 9 mirrored against a vertical axis; while
modern Hebrew uses the normal western script comma, which point "the
wrong way" (in my eyes, but as this is normal practice, it is in fact
OK).

In CJK we have a set of "full-width" punctuation marks in
addition to the "half-width" or Latin ones, and some
programs keep converting all full-width punctuation marks
to their half-width counterparts, thereby messing up the
alignment of characters.
There is also a set of vertical punctuation marks for
vertically running text...

How is it decided whether to run text horizontally or vertically? Arbitrarily?
Or are there rules that makes one un/less acceptable for a given putpose?

DKleinecke

2017-05-16 18:51:34 UTC

Post by Dingbat

True.
To make it more complicated, Arabic has its own special comma
character, shaped as a 9 mirrored against a vertical axis; while
modern Hebrew uses the normal western script comma, which point "the
wrong way" (in my eyes, but as this is normal practice, it is in fact
OK).

In CJK we have a set of "full-width" punctuation marks in
addition to the "half-width" or Latin ones, and some
programs keep converting all full-width punctuation marks
to their half-width counterparts, thereby messing up the
alignment of characters.
There is also a set of vertical punctuation marks for
vertically running text...

How is it decided whether to run text horizontally or vertically? Arbitrarily?
Or are there rules that makes one un/less acceptable for a given putpose?

Given a rectangular piece of "paper" there are four corners
where one might start and two directions at each corner that
one might go. That means there are eight possible layouts for a
succession of glyphs. This can be further complicated by varying
the move made when a traverse reaches the other side. Assuming
one doesn't want to jump across the empty spaces there are four
strategies making a total of 32 possible text layouts.

Only a few of the possibilities have actually ever been used in
real-life scripts. And there are still more ways to layout text -
like jamming everything into a box in an artistic way.

Peter T. Daniels

2017-05-16 20:31:39 UTC

True.
To make it more complicated, Arabic has its own special comma
character, shaped as a 9 mirrored against a vertical axis; while
modern Hebrew uses the normal western script comma, which point "the
wrong way" (in my eyes, but as this is normal practice, it is in fact
OK).

In CJK we have a set of "full-width" punctuation marks in
addition to the "half-width" or Latin ones, and some
programs keep converting all full-width punctuation marks
to their half-width counterparts, thereby messing up the
alignment of characters.
There is also a set of vertical punctuation marks for
vertically running text...

How is it decided whether to run text horizontally or vertically? Arbitrarily?
Or are there rules that makes one un/less acceptable for a given putpose?

Does that include outward spiral? (As on Mesopotamian incantation bowls, which
are in Aramaic, Mandaic, or Pehlevi.)

Post by DKleinecke
Only a few of the possibilities have actually ever been used in
real-life scripts. And there are still more ways to layout text -
like jamming everything into a box in an artistic way.

That would be the earliest Sumerian way.

DKleinecke

2017-05-16 23:05:57 UTC

Post by DKleinecke

Post by Dingbat

True.
To make it more complicated, Arabic has its own special comma
character, shaped as a 9 mirrored against a vertical axis; while
modern Hebrew uses the normal western script comma, which point "the
wrong way" (in my eyes, but as this is normal practice, it is in fact
OK).

In CJK we have a set of "full-width" punctuation marks in
addition to the "half-width" or Latin ones, and some
programs keep converting all full-width punctuation marks
to their half-width counterparts, thereby messing up the
alignment of characters.
There is also a set of vertical punctuation marks for
vertically running text...

How is it decided whether to run text horizontally or vertically? Arbitrarily?
Or are there rules that makes one un/less acceptable for a given putpose?

Does that include outward spiral? (As on Mesopotamian incantation bowls, which
are in Aramaic, Mandaic, or Pehlevi.)

That would be the earliest Sumerian way.

Aha, thanks. I forgot about spirals - at least four different
kinds. Mayan also seems to have a tendency to prefer an artistic
effect over readability.

Franz will be sad that the Phaistos Disk did not get enough
attention,

Franz Gnaedinger

2017-05-17 07:51:36 UTC

Post by DKleinecke
Aha, thanks. I forgot about spirals - at least four different
kinds. Mayan also seems to have a tendency to prefer an artistic
effect over readability.
Franz will be sad that the Phaistos Disk did not get enough
attention,

There is one more aspect, as the alphabet of 45 letters - or even 46 letters,
because the horizontal and vertical falcons have different phonetical values
in the decipherment of the Phaistos Disc by Derk Ohlenroth - also works for
pictures and allowed to compose a likeness of Tiryns (rosette of eight petals
in the center, emblem of Sseyr Zeus) and another of Elaia's grove at Phigalia
(bacon oven for Demeter-Elaia and river for Poseidon in the center, Poseidon
originally the god of rivers). Writing as picture can also be found for
example in Dada graphics. There are more ways of writing than horizontally
and vertically and in spirals.

Franz Gnaedinger

2017-05-17 08:00:20 UTC

Post by Franz Gnaedinger

Sorry for the typo, there is a baking oven in the center of the Elaia side
or disc (Phaistos Disc being a pair of clay discs baked together). A quite
similar baking oven was found in a garden sanctuary of the bird goddess
in Moldavia (if memory serves). The old tradition of the Vinca script
and garden sanctuaries would have survived in Elaia's grove at Phigalia,
which had a mighty influence on the Argolis via Eponymous Tiryns, Lord
Laertes the gardener in Homer's Odyssey, father of Odysseus. As for bacon,
pigs have been sacrificed to Demeter, as depicted in the entrance field
to Elaia's grove. Then there are bags with unwashed wool, also a sacrifice
for Demeter-Elaia. And portable beehives. We can not only read the spiral
texts but also glean plenty of information from the signs as pictograms.

Franz Gnaedinger

2017-05-18 06:33:26 UTC

Post by Franz Gnaedinger
Sorry for the typo, there is a baking oven in the center of the Elaia side
or disc (Phaistos Disc being a pair of clay discs baked together). A quite
similar baking oven was found in a garden sanctuary of the bird goddess
in Moldavia (if memory serves). The old tradition of the Vinca script
and garden sanctuaries would have survived in Elaia's grove at Phigalia,
which had a mighty influence on the Argolis via Eponymous Tiryns, Lord
Laertes the gardener in Homer's Odyssey, father of Odysseus. As for bacon,
pigs have been sacrificed to Demeter, as depicted in the entrance field
to Elaia's grove. Then there are bags with unwashed wool, also a sacrifice
for Demeter-Elaia. And portable beehives. We can not only read the spiral
texts but also glean plenty of information from the signs as pictograms.

The loudspeakers of the computer stations in my library being deactivated,
and having no earphones with me, watching the video makes no sense. But
I skimmed the comments and noticed a wheel-line and PIE *kwekwlos 'wheel'
which is very interesting in the context of the Phaistos Disc, Tiryns
side, rosette in the center.

CO OC LOP named the basic organization of a fortified settlement, attentive
mind CO right eye OC enveloping hedge or fence or palisade or wall LOP
(of a POL polis) - in the middle the ruler of the attentive mind CO, around
him guards of the open eyes OC along the enveloping palisade or wall LOP.
This concept is visualized by the Tiryns side or disc of the Phaistos Disc:
in the central field, next to the rosette of Seyr Zeus, the ruler of Tiryns,
Eponymous Tiryns, on his cheek a tattoo of two tiny circles, the upper one
indicating the sky, realm of Zeus, the lower one Tiryns, ruled by Eponymous
Tiryns in the name of Zeus; around him guards with their shields, watching
the gate, and looking over the palisade or wall. A variation of the shield
is the Argos eye, emblem of the watchful union of early cities in the
Argolis, a central dot surrounded by a circle of dots, on forehead and
cheeks and chin of the staring plaster head from Mycenae.

CO OC LOP named the cyclops, one-eyed giants, most famous Polyphem, Homeric
symbol of Troy VIIa. Achilles gained victory over Hector and dragged
his dead body three times around the walls of Troy, according to Eberhard
Zangger in order to break the spell of the cyclopic wall. This implies
that the protection provided by the wall had once been enforced by magic
incantations, analogous to the banning formulae of archaic power along
the margin of the Tiryns disc. I imagine a priest wheeling three times
around an early city (like Dimini, or the round cities in the Transural,
Nation of Cities, groundplans in some cases evoking a wheel with hub
and spokes and tyre) on a horse-pulled cart, and so it was easy to get
from CO OC LOP to Cyclops and *kwekwlos 'wheel' and Sanskrit cahkra 'wheel'
and English wheel.

You may remember that I locate the PIE belt from the region of the Göbekli
Tepe to Central Asia; the first IE homeland on the banks of the Amu Darya,
centered in the triangle of Termez and Kunduz and Kurgan T'upe (at the
northern base of the Alai Mountains where copper and tin were and are
associated in the same mines); the second IE homeland in the Uralic steppes
east of the Rha Volga; and the third IE homeland in the Pontic steppes
west of the Rha Volga, where Mallory assumes the origin of the wagon with
wheels. Rha, ancient name of the Volga, is akin to Rhea, mother of Zeus
(overcomer TYR Sseyr Sseus Zeus) and Poseidon (originally a god of rivers)
and Hades (god of the Underworld, originally also the mines, his dog
guarding the precious ores from the Alai Mountains).

I don't know what the video claims, but apparently some people are not
really satisfied with the current state of Paleo-linguistics.

Tak To

2017-05-22 00:32:11 UTC

Post by Dingbat

True.
To make it more complicated, Arabic has its own special comma
character, shaped as a 9 mirrored against a vertical axis; while
modern Hebrew uses the normal western script comma, which point "the
wrong way" (in my eyes, but as this is normal practice, it is in fact
OK).

In CJK we have a set of "full-width" punctuation marks in
addition to the "half-width" or Latin ones, and some
programs keep converting all full-width punctuation marks
to their half-width counterparts, thereby messing up the
alignment of characters.
There is also a set of vertical punctuation marks for
vertically running text...

How is it decided whether to run text horizontally or vertically? Arbitrarily?
Or are there rules that makes one un/less acceptable for a given putpose?

Traditionally all texts run vertically and the
columns go from right to left, except where the
vertical space (in shop signs, etc) limits the
columns to be just one character tall.

Publications with horizontal texts were introduced in
the last century to accommodate embedded foreign
scripts or scientific notations.

Nowadays the default format in Hong Kong and Taiwan
is vertical except for technical publications; unless
the author/publisher wants to effect a spiffy look, or
when they want to show compliance to the PRC (see
below).

The PRC made the horizontal format the standard at about
the same time they made the simplified characters the
standard -- they were necessary "advancements" for
modern age. Only books in/about Classical Chinese were
allowed to be done in the vertical format. Nowadays,
people are so accustomed to the horizontal format that
even those are done horizontally.

On the computer screen almost all texts are horizontal.

Ruud Harmsen

2017-05-22 07:04:54 UTC

Post by Tak To
Traditionally all texts run vertically and the
columns go from right to left, except where the
vertical space (in shop signs, etc) limits the
columns to be just one character tall.

But then the columns suddenly go from left to right?

--
Ruud Harmsen, http://rudhar.com

Tak To

2017-05-23 19:40:08 UTC

Post by Tak To
Traditionally all texts run vertically and the
columns go from right to left, except where the
vertical space (in shop signs, etc) limits the
columns to be just one character tall.

But then the columns suddenly go from left to right?

No. The columns/characters still go from right to
left. Traditionally, that is.

This picture of the logo of Tsinghua University
Loading Image...

shows both the tradition r-l order on the top and the
"modern" l-r order at the bottom.

Peter T. Daniels

2017-05-23 20:12:33 UTC

Post by Tak To
Traditionally all texts run vertically and the
columns go from right to left, except where the
vertical space (in shop signs, etc) limits the
columns to be just one character tall.

But then the columns suddenly go from left to right?

No. The columns/characters still go from right to
left. Traditionally, that is.
This picture of the logo of Tsinghua University
http://symlab.ust.hk/pic/logo/tsinghua.jpg
shows both the tradition r-l order on the top and the
"modern" l-r order at the bottom.

Is that licit because the upper script is formal and the lower is cursive?

Could it work like Egyptian, where you can run inscriptions from (exceptionally) left
to right in order to be symmetrical around (say) a doorway?

Tak To

2017-05-25 17:13:13 UTC

Post by Tak To
Traditionally all texts run vertically and the
columns go from right to left, except where the
vertical space (in shop signs, etc) limits the
columns to be just one character tall.

But then the columns suddenly go from left to right?

Is that licit because the upper script is formal and the lower is cursive?

The direction has nothing to do with the script. This is
the logo from another university which goes the other way.
Loading Image...

There was never a law about the direction in signs/seals
etc, but the PRC government took "rebellious attitudes"
quite seriously in the older days. Now r-l signs are
just weird.

That said, the PRC government seems to have given Tsinghua
University a lot of leeway. Note for example that their
official name is not "Qinghua" as it would have been
in pinyin. This is perhaps due to the fact that there
is also a Tsinghua University in Taiwan claiming to be
the true heir in spirit.

Post by Peter T. Daniels
Could it work like Egyptian, where you can run inscriptions from (exceptionally) left
to right in order to be symmetrical around (say) a doorway?

No. The characters always run counter-clockwise (5
o'clock to 7 o'clock) when the direction is right-to-left;
and clockwise when left-to-right.

Peter T. Daniels

2017-05-25 17:51:33 UTC

Post by Tak To
Traditionally all texts run vertically and the
columns go from right to left, except where the
vertical space (in shop signs, etc) limits the
columns to be just one character tall.

But then the columns suddenly go from left to right?

Is that licit because the upper script is formal and the lower is cursive?

The direction has nothing to do with the script. This is
the logo from another university which goes the other way.
http://img.mp.itc.cn/upload/20170329/358354337f29409f836ad7929b36646b_th.jpeg
There was never a law about the direction in signs/seals
etc, but the PRC government took "rebellious attitudes"
quite seriously in the older days. Now r-l signs are
just weird.
That said, the PRC government seems to have given Tsinghua
University a lot of leeway. Note for example that their
official name is not "Qinghua" as it would have been
in pinyin. This is perhaps due to the fact that there
is also a Tsinghua University in Taiwan claiming to be
the true heir in spirit.

Post by Peter T. Daniels
Could it work like Egyptian, where you can run inscriptions from (exceptionally) left
to right in order to be symmetrical around (say) a doorway?

No. The characters always run counter-clockwise (5
o'clock to 7 o'clock)

5 to 7 is by definition clockwise

Post by Tak To
when the direction is right-to-left;
and clockwise when left-to-right.

Tak To

2017-05-26 17:59:12 UTC

Post by Tak To
Traditionally all texts run vertically and the
columns go from right to left, except where the
vertical space (in shop signs, etc) limits the
columns to be just one character tall.

But then the columns suddenly go from left to right?

Is that licit because the upper script is formal and the lower is cursive?

Post by Peter T. Daniels
Could it work like Egyptian, where you can run inscriptions from (exceptionally) left
to right in order to be symmetrical around (say) a doorway?

No. The characters always run counter-clockwise (5
o'clock to 7 o'clock)

5 to 7 is by definition clockwise

Not if one goes via 12 o'clock.

Post by Tak To
when the direction is right-to-left;
and clockwise when left-to-right.

Ruud Harmsen

2017-04-27 07:39:51 UTC

Post by Tak To
In any case, given that the source HTML has neither a "dir" attribute
or a "direction" property[1], what makes you think that Firebox is
behaving correctly (and MS Word not) in the first place?
Incidentally, if one copies the entire table and not just the Hebrew
or Arabic text in the cell, the word order would be correct.

Just tried that (ctrl-A ctrl-C in
http://rudhar.com/sfreview/unigglen.htm, then ctrl-V in a fresh Word
document). Result: the Hebrew and Arabic is wrong.

--
Ruud Harmsen, http://rudhar.com

Tak To

2017-04-27 17:03:21 UTC

Just tried that (ctrl-A ctrl-C in
http://rudhar.com/sfreview/unigglen.htm, then ctrl-V in a fresh Word
document). Result: the Hebrew and Arabic is wrong.

As shown in the screen shot, word order on my computer is
correct, but text alignment in the paragraph/cell is not copied
over[1]. (I have enlarged the font size in the first column to make
the picture more legible; otherwise I have not touched the screen
shot.)
Loading Image...

[1] or copied over but overridden by the default style, which has
left alignment.

I can't tell you what happened at your computer, but I am running
MS Word 2007 and Firefox 53.0 on Windows 7.

Hope this helps.

Ruud Harmsen

2017-04-28 10:06:09 UTC

Post by Tak To
As shown in the screen shot, word order on my computer is
correct, but text alignment in the paragraph/cell is not copied
over[1]. (I have enlarged the font size in the first column to make
the picture more legible; otherwise I have not touched the screen
shot.)
http://i1132.photobucket.com/albums/m568/grandscholar/RTL%20in%20MS%20Word.jpg
[1] or copied over but overridden by the default style, which has
left alignment.
I can't tell you what happened at your computer, but I am running
MS Word 2007 and Firefox 53.0 on Windows 7.

In my case, character order, word order AND the alignment, are
automatically correct, in Firefox 52.0.2. (32 bits) under Windows 10
(64 bits).

In Word 2017 (after the mysterious fix), character and word order is
now correct, but the alignment is not.

--
Ruud Harmsen, http://rudhar.com

Athel Cornish-Bowden

2017-04-26 17:38:43 UTC

[ ... ]

Post by Ruud Harmsen
http://rudhar.com/lingtics/uniclnkl.htm
http://rudhar.com/cgi-bin/shunicod.cgi

Again links without explanation.

I just used them to check Mandaic and Tibetan. Well, but if you don't
want to know and don't want to learn anything, that's your free
choice. Freedom is paramount.

Although I've never had a need to type Tibetan I had a quick look to
see how well LaTeX could cope. I just inserted \usepackage{ctib} into
the preamble, and \tib in an arbitrary point later, and hey presto,
characters that look plausibly like Tibetan (to an inexpert eye) appear
in the output -- no fonts to be installed. It just works, and ctib is
designed to allow roman text to define what characters you want. It
could hardly be easier.

I have used the Greek equivalent, called polutonikongreek, and it
produces Greek words with accents and breathings without difficulty.
For example, I needed to include the word aitia in a recent paper (with
smooth breathing on the first iota, acute accent on the second) and it
was sufficient to put

\RequirePackage[english,polutonikogreek]{babel}%
\newcommand{\gk}[1]{\selectlanguage{polutonikogreek}#1\selectlanguage{english}}

in the preamble and \gk{a>it'ia} in the text. If you want to type a lot
of Greek text there are better ways of doing it (especially if you have
a Greek keyboard), but for just inserting the odd word here and there
it could hardly be easier.

There are similar packages for Arabic, Hebrew, Chinese (with Pinyin
input), and doubtless others.

--
athel

Ruud Harmsen

2017-04-26 19:27:56 UTC

Wed, 26 Apr 2017 19:38:43 +0200: Athel Cornish-Bowden

Post by Athel Cornish-Bowden
Although I've never had a need to type Tibetan

There's even a Wikipedia in Tibetan. 5th link in the blocklet at left
is "Random Article":
https://bo.wikipedia.org/wiki/%E0%BD%80%E0%BC%8B%E0%BD%98%E0%BC%8B%E0%BD%A3%E0%BC%8B%E0%BD%A2%E0%BD%80%E0%BE%B4%E0%BD%B2%E0%BC%8B%E0%BD%8F%E0%BC%8B

--
Ruud Harmsen, http://rudhar.com

Ruud Harmsen

2017-04-26 19:36:55 UTC

Post by Athel Cornish-Bowden

Wed, 26 Apr 2017 19:38:43 +0200: Athel Cornish-Bowden

Post by Ruud Harmsen
http://rudhar.com/cgi-bin/shunicod.cgi

I can do that with HTML entities too. Actually, that's how I do it in
http://rudhar.com/cgi-bin/shunicod.cgi . It's a program that sends
HTML.

--
Ruud Harmsen, http://rudhar.com

Ruud Harmsen

2017-04-26 19:39:33 UTC

Wed, 26 Apr 2017 19:38:43 +0200: Athel Cornish-Bowden

Post by Athel Cornish-Bowden
I have used the Greek equivalent, called polutonikongreek, and it
produces Greek words with accents and breathings without difficulty.
For example, I needed to include the word aitia in a recent paper (with
smooth breathing on the first iota, acute accent on the second) and it
was sufficient to put
\RequirePackage[english,polutonikogreek]{babel}%
\newcommand{\gk}[1]{\selectlanguage{polutonikogreek}#1\selectlanguage{english}}
in the preamble and \gk{a>it'ia} in the text. If you want to type a lot
of Greek text there are better ways of doing it (especially if you have
a Greek keyboard), but for just inserting the odd word here and there
it could hardly be easier.

This too can be done with Unicode, apparently, seeing this:
https://en.wikipedia.org/wiki/Ancient_Greek#Sample_texts

--
Ruud Harmsen, http://rudhar.com

Arnaud Fournet

2017-04-24 20:43:42 UTC