Subj : UTF-8 nodelist report
To : Sergey Dorofeev
From : Michiel van der Vlist
Date : Mon Mar 10 2025 12:16 pm
Hello Sergey,
On Monday March 10 2025 03:07, you wrote to me:
SD> About question mark, is it valid symbol for sysop name? Can somebody
SD> name himself as "???? ???? ????" for example?
Onde Fidonet sysop can ask more question than 10 FTSC members can answer. So I pass.
MV>> Very bad idea. Having more than one encoding within the same file
MV>> is a bad idea anyway, not just for the nodelist but for ANY text
MV>> file.
SD> Agree, thought it is technically possible, but it will kill 99%
SD> programmers.
There is no generally agreed mechanism to change the encoding in the middle of a text file. Is say: don't even try it.
MV>> Are you serious? You really still want every back alley in
MV>> Fidonet to have its own 8 bit encoding? With all the forward and
MV>> backward re-encoding and other limitations? C'mon.. That's chaos!
MV>> Unicode was invented for the very purpose of getting rid of all
MV>> this codepage shit.
SD> That's the thing that I do to keep all messages on my node in UTF-8...
SD> without stable success though.
OK. he message base however is something different. The best may be to archive them "as is". Or not, I don't know. But now we are no discussing the nodelist.
MV>> Why do you think Microsoft went full Unicode internally? Three
MV>> decades ago. Why do you think 99% of what is on the web is UTF-8?
MV>> To get rid of the mess of all the hundreds of 8 bit encodings
MV>> that floated around!
SD> There was two ways... make encoding a part of type, as if you use int
SD> for range, how to know, is it in miles or kilometers... or set the one
SD> and only encoding. Second way was chosen.
So let us stick to that choice.
MV>> Nah, as far as the nodelist goes, it is either just ASCII or
MV>> UTF-8. No more codepage shit.
SD> May be it is duty of ZC, to drop invalid segment, and not "help" sysop
SD> to publish errors, replacing them with any mark.
I say it is the resposibiliy of the RCs to submit an error free segment to the ZC. But let's not make a mauntain out of a mole hil. We have ONE RC submitting a segment that causes ONE line in the nodlist to be in error. An error that is largely cosmetic.
