Jeffrey Sabarese

hi there. thank you for your time!

i very much like your composition at
http://www.howtocreate.co.uk/sidehtmlentity.html

i find it very useful. i have never reproduced it, but i have saved it
locally for the purpose of quick-access on localhost, and in doing so, i've
discovered some things about Character Encodings, and the behaviour of IDE's
and Browsers vs. the declared Character Encoding.

First of all, i do not claim to be an expert on this subject, just to be
clear. :)
I realize that your entities page (in my awareness) does not have a declared
Character Set, so any time i have saved it, i have used the UTF-8 character
encoding, which does work fine, and i've also modified the doctype to XHTML
Transitional, UTF-8, which also works-- HOWEVER, if there is anything "out
of line", the character glyphs do INDEED become "skewed". I found this a
curious fact, so i tried to figure out what i had done "wrong" to make the
page NOT allow the browser to properly decode the entities as intended.

** MY INQUIRY **
To be honest, i am still unclear on the issue as a whole, however I have at
least determined a certain method for getting it "right"-- but it is
dependent upon the software i'm using ([brand 1]). For
example, using [brand 2] (and other softwares such as [brand 3]),
i've saved the page, declared the Character Set as UTF-8, and opened the
page to see that the decoded glyphs are not right. Yet, i've done precisely
the same w/ [brand 1] and it "works right".

PLEASE DON'T MISTAKE THIS FOR AN ENDORSEMENT! but a curiosity:
I wonder: what might it be about the different software; what is it that the
software does to the file "behind the scense" which causes it to remain
"unmodified", and therefore encoded properly? It is my guess that, for
example, [brand 3] must be adding or in some way "pre"-decoding
(while saving) the entities, and then re-encoding them-- improperly. It is
quite puzzling, and compelling for me. I wonder what are your own thoughts
on this?

i hope this is O.K. with you. i promise that i use it only locally, and have
to intent to publish it-- however, i will remove my saved copies if you
request i do so.

P.S. My web site where i reference howToCreate.co.uk:
[URL]
:)

thank you so much!
-J.S.

Tarquin,

Thank you for your prompt reply, and for the thoughtful responses to
my various inquiries.
this is the last i will contact you, unsolicited. thank you for your
understanding.

You might say i've conducted a bit of an experiment this morning,
inspired by your reply.

Curious what would happen in a variety of situations, i archived in
RAR format and attached to this mail a series of files which were
saved from the source-view of different Browsers, and from code
manipulation (or not) of the "htmlentities" file in different Editors.
(the file-names themselves should tell the tale)

What i hope to gain from this second mail, Sir, is a better
understanding of how i should go about coding in UTF-8, how to avoid
the pitfalls of non-compliant software, and-- rather than a "trial and
error" experiment as i did today, to simply "Know" for reasons of
sofware settings and features, which software supports, and which will
NOT support coding for i18n.

please see below, and the attached...

>> I realize that your entities page (in my awareness) does not have a
>> declared Character Set
>
> It does. I use the Content-Type header as part of the HTTP
> communication. The correct character set is utf-8.

Forgive me, i do not mean to argue the point, but could you tell me
which line this is? am i going mad? i do not see it. perhaps you've
written it in a way which i am unfamiliar?

>> For example, using [brand 2] (and other softwares such as [brand 3]...SNIP
>
> The base of UTF-8 is compatible with ascii, but the higher ranges are
> completely incompatible.....SNIP...it saves the modified broken output,
> but in whatever way it interpreted them, instead of in the same bit
> sequence as they were in when you opened it. So it ends up broken.

THANK YOU! it's the "upper UTF-8 element which was puzzling. now it
makes sense. thanks you for that understanding! the question now
remains: which editors to use? :)

> In the case of the [brand 3], I do not know exactly what it does,
> but I suspect it does the same thing. It may well also remove my
> entities from the source and replace them with exact characters in
> 8859-1, which confuses it even more when you tell the browser to
> interpret it as utf-8.

meaning, when i added the meta-tag w/ the UTF-8 ?? or simply setting
the browser to read it as such... or?

> You need to use an editor that understands utf-8 and either recognises
> the file as that, or allows you to tell it what the file is. I use an
> editor that does that. Your other editor seems to be able to do this too.

If you don't mind, i'd like to know what you recommend for such compatibilities.
i think i understand most of it now.

One curiosity remains: What if, for example, [brand 3] "ruins" the
file by putting / warping the encodings. is there any way to "un-do"
this process, or is the file essentially ruined?
How can i be sure that when i "view source" that the encodings are not
destroyed at that very moment?

thanks very much for your time!
Jeffrey Sabarese