Hello,
Summary: One may produce Hebrew pages with output in either ISO-8859-8 or
ISO-8859-8-i, but SHOULD serve them as ISO-8859-8-i over the web.
Strictly speaking ISO-8859-8 and ISO-8859-8-i
are the same encoding. Namely, they assign identical bytes to the same
characters.
Howver due to historical reasons one should not use the notation ISO-8859-8
for HTML documents over the WWW as many browsers assume that pages in that
encoding are
encoded using the "visual method" rather with Unicode's implicit
directionality ("the logical method").
As for generating HTML pages using XSLT or some other SGML/XML
mechanism, one may configure the output encoding to be ISO-8859-8,
even when the pages are written in the logical method.
But they should be served via HTTP with the notation:
Content-type: text/html;charset=iso-8859-8-i
Hope this clarifies the issue.
Please, do not hesitate to ask for more clrafications, if needed.
Nir.
----- Original Message -----
From: "Shlomi Loubaton" <[EMAIL PROTECTED]>
To: "Gabor Hojtsy" <[EMAIL PROTECTED]>
Cc: "phpdoc" <[EMAIL PROTECTED]>; <[EMAIL PROTECTED]>
Sent: Monday, January 14, 2002 12:55 PM
Subject: Re: [PHP-DOC] Re: hebrew translation
>
> > > OK, for the SP encoding (we need this for HTML generation),
> > > these seems to be good:
> > >
> > > is8859-n or iso-8859-n
> > > n can be any single digit other than 0. Each character in the
> > > repertoire of ISO 8859-n is represented by a single byte.
> > >
> > > Now the question to Shlomi is that what -i means in the encoding.
> > > Hope it is not needed :)
> > >
> > > It is is, please look into your jade distributions charset.htm
> > > file, and choose the best encoding from there for the "he" files.
> >
> > It seems I write too many emails before searching for the solution :)
> > I found this test site: http://www.nirdagan.com/hebrew/characters/:
> >
> > | Test page encoded in ISO-8859-8-i. This is an identical encoding to
> > | ISO-8859-8. It was introduced by RFC 1556 to distinguish between
> > | logical and visual storing order. As far as HTML and XML are concerned
> > | they are identical. In HTML all document are stored logically.
> > | See overview of standards for details.
> >
> > The overview of standards page
(http://www.nirdagan.com/hebrew/standards)
> > points out, that using of iso-8859-8 is adviced for backward
> > compatibility, and HTML even do not care about embedded
> > directionality info in encodings, so from now on, we will use
> > iso-8859-8 for Hebrew :)) This simplifies things a bit :))
> >
> > Goba
> >
> >
>
> I'm not sure that we should choose iso-8859-8 since it's a very
problematic
> encoding. I know for example that you have problems with it when you break
> lines, mix english and hebrew text and it looks different in different
> browsers.
> and it doesn't solve RightToLeft problem since i still have to insert <p
> align=right> or something like that - in order to set text alignment. and
> when translators will work using this encoding, they'll have to write
> everything backwards .. that's too hard to translate the whole manual this
> way.
>
> Maybe i should ask Nir Dagan for his opinion.
>
> Regards
> Shlomi Loubaton
> [EMAIL PROTECTED]
>
>
