page encoding question

[Tony Vella]

I am preparing a series of philatelic html pages (lots of text and a few
scans of stamps) which will include alpha-characters (accents) in Italian,
French, Spanish, Portuguese and Danish. The pages I have finished in draft
form so far I have encoded UTF-8 but I have just been told that 99% of the
world will not be able to read them and that I should go through all the
pages and re-encode them "western european - windows (1252)". I guess what
I would like to know is what encoding would be most effective for these
particular languages. Any advice and pointers will be appreciated.

 

[David Dorward]

I am preparing a series of philatelic html pages (lots of text and a few
scans of stamps) which will include alpha-characters (accents) in Italian,
French, Spanish, Portuguese and Danish. The pages I have finished in draft
form so far I have encoded UTF-8 but I have just been told that 99% of the
world will not be able to read them

That is rubbish. UTF-8 is very well supported (so much so, that I can't
remember the last time I came across a system that couldn't handle it).

 

[Dan]

I am preparing a series of philatelic html pages (lots of text and a few
scans of stamps) which will include alpha-characters (accents) in Italian,
French, Spanish, Portuguese and Danish. The pages I have finished in draft
form so far I have encoded UTF-8 but I have just been told that 99% of the
world will not be able to read them and that I should go through all the
pages and re-encode them "western european - windows (1252)". I guess what
I would like to know is what encoding would be most effective for these
particular languages. Any advice and pointers will be appreciated.

While UTF-8 is actually very widely supported, and thus there's no
reason to change your encoding from this (if your server sends a proper
content-type header indicating the encoding), the Western European
languages you are using should work all right in the standard Western
encoding iso-8859-1 as well. Avoid windows-1252; it's a proprietary
Microsoft set.

 

[Luigi Donatello Asero]

While UTF-8 is actually very widely supported, and thus there's no
reason to change your encoding from this (if your server sends a proper
content-type header indicating the encoding), the Western European
languages you are using should work all right in the standard Western
encoding iso-8859-1 as well. Avoid windows-1252; it's a proprietary
Microsoft set.

I am using
iso-8859-1 at the moment but I am going to change it into UTF-8 to add the
pages in Russian and Chinese
(just now I have little in these languages)

 

[Jukka K. Korpela]

I am preparing a series of philatelic html pages (lots of text and a few
scans of stamps) which will include alpha-characters (accents) in Italian,
French, Spanish, Portuguese and Danish.

They are all covered by the ISO-8859-1 encoding, except for some punctuation
marks and letters like the oe ligature. If you use windows-1252, you get the
punctuation marks and the ligature, too.

The pages I have finished in draft
form so far I have encoded UTF-8 but I have just been told that 99% of the
world will not be able to read them

Nonsense. More probably, 99 % of the WWW users _are_ able to read them. Well,
let's say 97.6 %. After all, 96,3 % of all percentages have just been made
up, and the remaining 4,7 % have been miscalculated.

and that I should go through all the
pages and re-encode them "western european - windows (1252)".

I wouldn't do that at this point, unless you have good tools that do such
things for you with minimal effort.

I guess what
I would like to know is what encoding would be most effective for these
particular languages.

If you were just about to start the project, I would recommend ISO-8859-1 (or
windows-1252 if you need those extras) - not because of wider browser
coverage (though there is a _small_ improvement to be gained there) but
because those encodings are somewhat more efficient (one byte per character,
whereas UTF-8 uses two bytes for some of the characters you'd use).

UTF-8 is certainly simpler in the future if you'll ever need to add
characters in other languages.

 

[Alan J. Flavell]

That is rubbish.
Agreed

UTF-8 is very well supported (so much so, that I can't remember the
last time I came across a system that couldn't handle it).

Broad agreement with that, but there are exceptions...

Well, Netscape 4.* versions do a pretty good job of *rendering* utf-8,
but do keep in mind that, if any form submission is required, then NN4
badly mangles utf-8. Whether it's worth understanding how to
implement a workaround for that old zombie is debatable, of course:
I'm just mentioning that it's not without a problem.

cheers

(The original WebTV is also hopeless at rendering anything other than
a subset of Windows-1252, but ho hum.)