[R] [OT] "_" inserted in postings

[(Ted Harding)]

On 08-Sep-09 10:40:18, Duncan Murdoch wrote:
> On 08/09/2009 6:09 AM, (Ted Harding) wrote:
>> Sorry if this is too "OT", but there is a particular relevance to
>> postings to R-help.
>> 
>> Of recent times, I have received several postings via R-help (and some
>> other mailing-lists) in which the "_" character is inserted where,
>> presumably, a space, " ", was intended. An example (received this
>> morning) is below, which (from the headers) was originally sent
>> through Yahoo web-mail. This is as seen when I read it.
>> [ From: FMH <kagba2006 at yahoo.com>
>>   Subject: [R] Derivative of nonparametric curve
>>   Date: Tue, 8 Sep 2009 02:07:10 -0700 (PDT) ]
>> 
>> ================================================
>> Dear All,
>> 
>> I'm looking for_a way on computing the derivative of first and second
>> order_of a smoothing curve produced by a nonprametric regression. For
>> instance, if we run the R script below, a smooth nonparametric
>> regression
>> curve is produced. 
>> 
>> provide.data(trawl)
>> Zone92__ <- (Year == 0 & Zone == 1)
>> Position <- cbind(Longitude - 143, Latitude)
>> dimnames(Position)[[2]][1] <- "Longitude - 143"
>> sm.regression(Longitude, Score1, method = "aicc", col = "red",_ model
>> =
>> "linear")
>> 
>> Could someone please give some hints_on the way to_find the
>> derivative_on
>> the curve at_some points ?
>> 
>> Thank you.
>> Kagba
>> ================================================
>> 
>> In the message headers I see:
>> 
>>   Content-Type: text/plain; charset="iso-8859-1"
>>   Content-Transfer-Encoding: quoted-printable
>> 
>> and, on inspection of the message text in the "inbox" folder (i.e. as
>> it was delivered to me), I see that each and every occurrence of the
>> "_" is represented there as the three successive characters "=A0",
>> i.e. a "quoted-printable" code.
>> 
>> Normally, I can happily ignore this kind of thing when it just occurs
>> in text. But since, as in the above message, it can also be
>> interpolated
>> into R code, this could cause unnecessary inconvenience for people
>> who want to test the code which people post to R-help. Ditto if
>> someone
>> should post code (such as the above) as a solution to someone else's
>> problem: As received, it just would not work!
>> 
>> Comment: I have for long time been under the impression, now
>> apparently
>> a delusion, that the abominable quoted-printable had found its due
>> final
>> resting-place in the Museum of Dishonorable Obsolescence; apparently
>> not!
>> 
>> Also, if people are using web-mailers (Yahoo or other) that wantonly
>> insert this kind of rubbish, they should look into the possibility of
>> either changing the configuration under which they post their mails
>> (if possible), or mailing via a different agent.
>> 
>> The "quoted-printable" aspect may be a red herring, since I also see
>> that the "=" signs in the code are represented as "=3D", as is forced
>> in "quoted-printable"; but they have been rendered correctly in the
>> text as seen (and copied). So it may just be due to substitution of
>> "_" for " " on the ,art of the mailer.
>> 
>> What do others think?
> 
> I don't see the underscores in that posting, but I do see this as the 
> last line in the headers:
> 
> X-MIME-Autoconverted: from quoted-printable to 8bit by 
> fisher.stats.uwo.ca id n8898BkC031633
> 
> (fisher.stats.uwo.ca is the server that receives my email).
> 
> So it looks as though whatever is doing the conversion on your system 
> isn't doing it as well as it should.
> 
> Duncan Murdoch

Thanks for this. No such conversion was performed in my case: it was
delivered as-is, i.e. in the original quoted-printable (QP). The
destination to which it was originally delivered (manchester.ac.uk)
apparently does nothing about re-encoding it. My local mailer handles
the decoding and rendering of encoded content.

The QP code "=A0" in the source
  Content-Type: text/plain; charset="iso-8859-1"
corresponds to "Non-Breakable Space" (NBSP), so presumably my mailer
displays this as "_" to distinguish it from the basic ASCII space (SP)
(ASCII code 32, QP code "=20"). (Note that the QP elements "=3D" in the
original were correctly rendered as "=", so the decoding indeed took
place).

This still leaves open the issue that what was presumably a simple
ASCII space SP when originally entered, got changed to NBSP somehow
in the process of being sent!

Thanks for the comment,
Ted.

--------------------------------------------------------------------
E-Mail: (Ted Harding) <Ted.Harding at manchester.ac.uk>
Fax-to-email: +44 (0)870 094 0861
Date: 08-Sep-09                                       Time: 12:46:42
------------------------------ XFMail ------------------------------