On Sat, 5 Oct 2013 17:20:38 -0400, John Gilmore wrote:

>Bill,
>
>What you sent as an extract from my text arrived as
>
>>Consider the botched French-language text
>>
>>A Montréal, a la fin des années 80, . . .
>>
>>It should of course be
>>
>>A Montréal, a la fin des années 80 . . .
>>
>
>i.e., with good and bad text strings interchanged.

I'll take the blame for the second part being wrong. I replied to your original 
post using the listserv's web interface at 
https://listserv.ua.edu/cgi-bin/wa?A0=IBM-MAIN and when I told it to quote your 
message (clicking on the big double quote at the bottom), the result was that 
the pair of bytes that you used in the deliberately botched line (hex C3 and 
A9) were shown as the accented "e", having been interpreted as if they were 
UTF-8. The second representation of the line, which had the accented "e" in 
your original post, was now presented as a white question mark on a black 
diamond-shaped background, which is the Uncode Replacement Character U+FFFD as 
shown at http://www.charbase.com/fffd-unicode-replacement-character

Anyone with access to the listserv's web interface (well, maybe not with all 
browsers and OSes) can see what I'm talking about by starting to reply to your 
original post, quoting it, and looking at how the quoted text  has been  
transformed. Then back out without sending.

I decided to correct the quoted text that the web interface had modified, but I 
got it backwards. I left the first line unchanged, when I should have replaced 
the accented "e" with the original pair of bytes, and I replaced the U+FFFD in 
the second text string with the original pair of bytes, when I should have used 
an accented "e".

I should add that in quoting your reply for this post, which is supposed to 
show the good and bad text strings interchanged, the web interface, consistent 
in its perversity, now shows the first instance of the string with the U+FFFD, 
and the second instance with the accented "e". I have fixed up both instances 
before sending, so with luck it will show the text string interchanged, and it 
will be an accurate quote of your reply, with my mistake.

Bill

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN

Reply via email to