User Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; it; rv:1.9.2.24) Gecko/20111103 Firefox/3.6.24 ( .NET CLR 3.5.30729) Build ID: 20111103063747 Steps to reproduce: install Thunderbird Actual results: The default encoding was not UTF-8, and not even UTF-16, and ... not Unicode! Expected results: Hello all. Everybody experienced problems with accented letters or different alfabeths, expecially who speaks more than one language or has foreign friends who want to give their phisical address. It's true, those problems are always less, anyway i think that a program like Thunderbird should have as default encoding UTF-8 just after install. As we know, UTF-8 allows to show all the alphabets in the world at the same time (i could write in hundred of language in the same document), it's suggested by W3C (or even standard of W3C, with UTF-16). Let's take the e-mail case. The condition to see correctly the caracters is that both the sender and receiver use UTF-8 (or compatible) as encoding. UTF-8 is an Unicode encoding; Unicode exists since ~ 20 years (!!) and UTF-8 since a little bit later. Anyway not all emails are received as they should. As free program, Thunderbird should offer the best: UTF-8 as default (so people don't need to learn about it to read emails in all language); and if somebody wants a monolingual or "few-lingual" encoding, can change it in the settings. This would be a big help for many people who don't have technical knowledge, and a big step in the direction of a world in which al language can circulate without problems (this should be just after Unicode, in '90s, so after 20 years we are very late! lets's do it ASAP! ;) ).

(In reply to UTF8-unicode from comment #0) > Steps to reproduce: > install Thunderbird Those are not steps to reproduce. Rather please provide where you found out in the settings about UTF-8 not being the default. > > Actual results: > The default encoding was not UTF-8, and not even UTF-16, and ... not Unicode! So what is it? "Actual results" should mention what the results are, not what they are NOT. :)

(In reply to Andre Klapper from comment #1) > (In reply to UTF8-unicode from comment #0) > > Steps to reproduce: > > install Thunderbird > > Those are not steps to reproduce. Rather please provide where you found out > in the settings about UTF-8 not being the default. > > > > > Actual results: > > The default encoding was not UTF-8, and not even UTF-16, and ... not Unicode! > > So what is it? "Actual results" should mention what the results are, not > what they are NOT. :) Yes, i agree, but i didn't know what to write in those textboxes. It's not a real bug, but something that gives problems (so quite annoying, like a bug). I installed Thunderbird on Windows XP, and in the settings the default encoding was not UTF-8, but a regional encoding, that will cause problems to people who use a different alfabet, or to me when i receive emails with letters from other languages

Starting the en-US version of Thunderbird 8.0 on Windows 7 with a new profile gives me ISO-8859-1 as the default encoding. Per bug 410333 a message will be sent in UTF-8 if there is at least one character which cannot be represented in the default character set (the latter is defined by the locale used, thus would be different if you install a build in a language other than en-US). There is an exception due to UTF-8 not being well handled in certain locales, that's bug 448842. Making UTF-8 the default encoding was won't-fixed in bug 224391. Unless any of the rationales discussed there (or in the other bugs) do no longer apply or warrant reconsideration, this bug should be closed as a duplicate.

(In reply to rsx11m from comment #3) > Starting the en-US version of Thunderbird 8.0 on Windows 7 with a new > profile gives me ISO-8859-1 as the default encoding. ... > Making UTF-8 the default encoding was won't-fixed in bug 224391. Unless any > of the rationales discussed there (or in the other bugs) do no longer apply > or warrant reconsideration, this bug should be closed as a duplicate. I hope that after 4 years UTF-8 is more accepted by all the mail providers (this is one of the point that caused the closing of the bug 224391). For example, few years ago I used Libero mail (an italian one). I left it and created a Yahoo account just to see non-english and non-italian characters in the same email. In fact, when i need to send some box with libero, i had to ask people to send me a word document with their address. Now i know that even Libero recently accepts UTF-8. Else will loose other people as it lost me. If Thunderbird, one of the most used email clients uses UTF-8 as default, many people will use UTF-8 without even need to change any option or without any need of technical knowledge more and more people will have less and less problems with characters. If in the Thunderbird start page you would inform people about this new feature, will be a good way. -------- (Anyway, it's just a shame that somebody cannot send some simple text to some customer/provider/friend etc. and in the internet era, this will not arrive. I mean, a picture will be safe, but a word will not... are we in 2011?)

(In reply to UTF8-unicode from comment #4) > (Anyway, it's just a shame that somebody cannot send some simple text to some > customer/provider/friend etc. and in the internet era, this will not arrive. I don't get that part. As written in comment 3 already, "a message will be sent in UTF-8 if there is at least one character which cannot be represented in the default character set".

(In reply to Andre Klapper from comment #5) > (In reply to UTF8-unicode from comment #4) > > (Anyway, it's just a shame that somebody cannot send some simple text to some > > customer/provider/friend etc. and in the internet era, this will not arrive. > > I don't get that part. As written in comment 3 already, "a message will be > sent in UTF-8 if there is at least one character which cannot be represented > in the default character set". This is not the total solution. If a Chinese guy writes to me (italian) he will use characters that will be all in her default character set, that is (by default in Thunderbird) different than UTF-8. So i will not be able to see them if in my client i don't his character set. The setting of UTF-8 as default is just to "educate" people to have this encoding, even if they don't know about that. One world, many languages, one encoding. It seems so simple and no-damage, and back compatibility (see the next example...). Since pure ASCII is 7 bits (one bit was for checking, later used for national languages different than English), UTF-8 for English is exactly equal than ASCII. This is an UTF-8 Byte: FXXXXXXX If F == 0, then XXXXXXX is read as pure ASCII; else if F==1, then XXXXXXX is concatenated with the following byte to get a non-ASCII character. For English speakers there is no-problem; US people can even keep thinking that their US-ASCII is international (they can read as UTF-8 even the ASCII text wrote 35 years ago...). And in addition, a great advantage will be for all the rest of world languages. Nos, people still cannot send their words and thinking in email (it's a shame, as i told), but they can send pictures anything else. Anythings except their words in a email message.

(In reply to UTF8-unicode from comment #4) > If in the Thunderbird start page you would inform people about this new > feature, will be a good way. Which new feature are you referring to? Sending a message in UTF-8 if it doesn't fit into the default encoding has been in there much longer than bug 410333, which only removed the (somewhat confusing) dialog to confirm if to do so.

(In reply to UTF8-unicode from comment #6) > This is not the total solution. If a Chinese guy writes to me (italian) he > will use characters that will be all in her default character set, that is > (by default in Thunderbird) different than UTF-8. So i will not be able to > see them if in my client i don't his character set. Similar to comment #5, I don't see the problem. If a message is encoded in a different character set, Thunderbird will display it in that character set as specified in the message headers, /unless/ you chose the option to enforce a display in the selected default encoding in Thunderbird. It's the receiving client's task to interpret and display the encoding correctly.

(In reply to rsx11m from comment #8) > (In reply to UTF8-unicode from comment #6) > > This is not the total solution. If a Chinese guy writes to me (italian) he > > will use characters that will be all in her default character set, that is > > (by default in Thunderbird) different than UTF-8. So i will not be able to > > see them if in my client i don't his character set. > > Similar to comment #5, I don't see the problem. If a message is encoded in a > different character set, Thunderbird will display it in that character set > as specified in the message headers, /unless/ you chose the option to > enforce a display in the selected default encoding in Thunderbird. It's the > receiving client's task to interpret and display the encoding correctly. That's a good thing. But 1) still seems that the national encodings are just a useless weight of the past... And 2) anyway this works with only-Thunderbird client send-receiver. But if all the clients in the world would have the same encoding, all the editors would use the same encoding... Of course, let's start by Thunderbird ;-)

(In reply to UTF8-unicode from comment #9) > But 1) still seems that the national encodings are just > a useless weight of the past... That's in essence an internationalization problem and to which extent the UTF-8 encoding has been adopted (already). A look at http://mxr.mozilla.org/l10n-central/search?string=intl.charset.default shows that some locales already use UTF-8 as their default encoding. Thus, the system seems to work as is. The main question is if the en-US default locale should be switched to UTF-8 to encourage use of that generic encoding in others. > And 2) anyway this works with only-Thunderbird client send-receiver. Different encodings should be supported by all standard-compliant clients, where not all encodings may be actually installed by default on a specific system.

FYI, intl.default.charset is not a default charset for mail. https://mxr.mozilla.org/l10n-central/search?string=mailnews.*_default_charset&regexp=on

Thanks and sorry for mixing that up. Yet quite a few UTF-8 defaults there.

(In reply to rsx11m from comment #10) >The main question is if the en-US default locale > should be switched to UTF-8 to encourage use of that generic encoding in > others. Why not? One world, one family, same problems, same solutions :) ... unique encoding for the humanity. User Masatoshi Kimura in comment #11 confirms that still not everything is so UTF-8. In the first replies users have written about similar bugs since few years, so it's a need that many people feel (other are just used to the problems and think there is no solution). UTF-8 should be the future but it newer becomes present...(?) free software as thunderbird, firefox etc. should be an example for all. Linux is already UTF-8 based, and also the other main O.S..

(In reply to UTF8-unicode from comment #13) > unique encoding for the humanity. > UTF-8 should be the future but it newer becomes present...(?) free software > as thunderbird, firefox etc. should be an example for all. Please stay technical and avoid platitudes. It really doesn't help convincing. Thanks.

(In reply to Andre Klapper from comment #14) > (In reply to UTF8-unicode from comment #13) > > unique encoding for the humanity. > > UTF-8 should be the future but it newer becomes present...(?) free software > > as thunderbird, firefox etc. should be an example for all. > > Please stay technical and avoid platitudes. It really doesn't help > convincing. Thanks. Ok, I understand. But from software engining point of view this is not so platitude as it can seem. I wrote with not technical words something that is technical, sorry for that. I had an experience with a firm that was able to convince 90% of its customers to use a standard for some files. It was a big help to have 90% of customers with the same standard, but they weren't 100%, so the source code in many parts contained checkings like: IF (standard) then ... ELSE IF (customer1) ... ELSE IF (customerN) So you can imagine, different bugs, different parsers, the new workers of the firm had to learn all the different formats and this meant more training etc. One format is always the optimal choice, when it is possible.

(In reply to [:Aureliano Buendía] from comment #16) > > *** This bug has been marked as a duplicate of bug 224391 *** Does it mean that: 1. the bug will be anyway evalued considering it together with the duplicated one or that 2. the problems with encoding will continue forever because it's not possible to re-consider a decisions token in the past (unless also the competitors of Thunderbird will do the same, but before) ? I think that the need of having the higest possibility to read at least own name is almost a people's rigth... For example a signature: Fëdor Michajlovič Dostoevskij Фёдор Михайлович Достоевский or the phonetica alphabet: fʲodər mʲɪˈxajləvʲɪtɕ dəstɐˈjɛfskʲɪj But i won't continue the discussion, even if a bit disappointed if the answer is 2.

[I mean, i won't continue in the sense that i won't bother anymore]