Forums Feature Suggestions Suggest making UTF-8 the default setting

In my experience the UTF-8 support in mIRC is developed enough and fixes more problems than it causes. The only "problem" really is that some luddites complain because they're too lazy to update their IRC clients. ;-) (However, I think that enough time has passed - it's been over a year since mIRC itself started supporting displaying UTF-8 - surely everybody's updated by now.)

I frequently chat in truly multilingual channels and it's very time consuming to instruct newcomers on how to make it possible to enter any character in mIRC. Most of the people don't even realize their characters aren't showing up correctly at the other end until they're told about it.

Therefore, I humbly propose that multibyte display and multibyte editbox are made the default settings along with UTF-8 set to "display and encode" for every new channel and query window.

Don't forget that there are other clients out there that do not support unicode at all. Having that as the default option isn't a good idea, imo. When you know you want to use it, then you know that you need to change settings (even if you don't know which ones). However, for everyone who doesn't need unicode, they are less likely to know that they need to change anything to make it show properly for others.

If it's troublesome explaining how to set it up correctly in your channel(s), then use a script to explain it through /msg, /notice, or even /dcc send'ing a file. You can have an on join message telling people to type !unicode or whatever the first time they join in order to get set up correctly. It's automated that way.

I think that leaving it off is a good idea until the majority of clients work with unicode and most people are using the latest versions.

Simple enough to just /utf-8 and tell them to get their act together.


Also, if they're using a client that doesn't support unicode or utf-8, then they need to get a better client ;P Seriously.. more things are uniform when one common encode type is used, and not having to use crappy encoding scheme's like shitjis instead of the one where you can type in all your languages at once is just better in the long run. So I second the "make utf-8 default" as it's just better.. and hell, remove shitjis support so people can get their act together and stop being lazy with poor excuses like "that chan only uses shitjis so i have to" but instead, if it was already set up for them.. they'd be just as fine and have to do less work, can't be hard for Khaled to just set a couple flags to 1 for those options as default.

That's the problem. Most people don't know they want to use it until they've wasted a lot of time trying to figure out why they are seeing garbage, or why people aren't responding to them (because the other people are seeing garbage as well).

My gut feeling says that there would be very few people wanting to switch *away* from UTF-8 were it the default. Although admittedly, those few people would be very vocal about it..


Onjoin messages are a significant annoyance to most people. It would be much more helpful if UTF-8 were the default, then I could just tell people to install mIRC and everything would work out of the box. I'm a firm believer in the zeroconf philosophy. Type in any language and boom, it works.

It doesn't have to be an on join message. That was an example. You can have the command listed in the topic, run from an ad in the channel, or just tell people to use it. Until/If the default is changed, that's your best option.

I still stand by the fact that until all major clients support Unicode, the default setting shouldn't be changed.

All the major clients already support UTF-8, infact, many supported it before mIRC did.

http://en.wikipedia.org/wiki/List_of_IRC_clients#Features

I've no idea how many of those use it by default, though.

Your 'ALL' is somewhat inaccurate... miranda, ircle, trillion, virc are all major clients that are not listed as supporting utf-8, infact those few are more popular than most of the others than do support it (i said most not all).

On a similar note, how many of the irc-bots support utf-8? (ie eggdrop, energymech)

That being said, i have UTF-8 enabled and i dont believe i have ever not been able to read what somebody has said. Under what circumstances would somebody with UTF-8 enabled not be able to read text that they would be able to read is it was disabled?

I believe Miranda IM has a patch available for UTF-8 support. At least some versions of Trillian/Trillian Pro support it too. As far as I know Visual IRC is not particularly popular, even if it is there hasn't been a release since January 2004. Given that it's open source if there hasn't been a new version in over three years I think it's fair to say that it's not very popular and that it's possible there may not even be another version.

Which leaves IRCle. Which is the only one of the four that is a major client really. A quick check of the version history shows that as of February 2006 Unicode/UTF-8 support is planned "for the very near future". So I guess it should be available soon. Then again if you read the IRCle forums you'll see Onno said that UTF-8 would be supported "later this year" in a post from August 2004.

I have no idea if any IRC bots handle Unicode/UTF-8. Believe me when I say I couldn't care less about them.

The UTF-8 encoding matches ASCII for codepage values 0-127 (ie. control chars, a-z, A-Z, 0-9, all common punctuation), so if you chat in English you won't notice the difference.

They don't have to. That's one of the great things about UTF-8.


The issue at hand is regarding whether mIRC should send UTF-8 by default; if it did, then a receiver would now fail to see the proper text if all of the following conditions are met:

• The receiving client does not support displaying of UTF-8 or has displaying of UTF-8 disabled;
• The text sent is using characters not in the ASCII 0-127 range (i.e. non-English language);
• The sender and receiver had the same codepage enabled originally (otherwise the receiver would have seen gibberish anyway).

Note that a person with UTF-8 enabled can always still properly see the text of a person with UTF-8 disabled because of the backwards compatibility for non-UTF-8 text.

Personally, I support this suggestion, although it may be a little too early; there will also be many unexpected side effects that people will start having to deal with (e.g. the fact that joining channels with non-UTF8 names but ASCII 128-255 characters in them would become surprisingly hard), so this change would have quite some impact - but eventually this is definitely the way to go.

The second best thing (at least for now) IMO would be a "master switch" in the form of a button in the Options dialog, or a command or so, that allows you to enable all settings necessary for full use of UTF-8 at once. Right now, enabling all the necessary settings one by one (in the options, and then for every window type etc) is just way too cumbersome.

A simple command for setting a window / window type to unicode or utf-8 would be great for the time being, ie:
/commadsnd <-dsacntx> <unicode|utf-8|etc> [<CURRENT|network>[,<network2>,etc]] [<CURRENT|type>[,<type2>,etc]]
-d display
-s send
-a all
-c current window
-n network
-t type
-x makes it only untill the window is closed
type: status|etc
e.g. /command -dsa utf-8, would set send and display for all windows.
or perhaps /command -dsc utf-8, would send and display utf-8 only for the current window

i believe this would simplify how people could configure their scripts, and in turn explain the command they need to do to enable seeing and or typing using the correct encoding

cheers