trouble with the "Submit Comments or New Sequences" web page

N. J. A. Sloane njas at research.att.com
Sat Nov 5 23:14:25 CET 2005


Something really horrible is going on.

When you fill out that web page, it goes to a program
in the cgi bin that converts what you typed into an
email message, and mails it to me.

That seems to be the preferred way to get things through the firewall.

The bad thing that is happening is this:  Don Reble discovered
that certain harmless lines are considered to be a virus by
the AT&T mail program at mail-dark.research.att.com.

When it sees one of these lines, it rejects the email with a
message saying:

> 550 Error: Your email looked like it contained a virus.
> 554 5.0.0 Service unavailable

But the cgi program doesn't see this, and so the message
vanishes, and your input is lost.

An example of something that appears to be a virus is
the line:

%D A083378 W. Huerlimann, Integer powers and Benford's law, preprint, 2003, available at www.mathpreprints.com/math/Preprint/werner.huerlimann/20030603/1 or at www.geocities.com/hurlimann53 (list of publications)

The moral of the story is that if you send me (or have sent me in
the past) a comment or new sequence that never made it into the
OEIS, it may be because it never reached me!

I wll try to see if this can be fixed.

Neil

PS I don't know what it is about that line that looks bad.
If anyone wants to run some tests, send in a Comment with a line
like that, and put in another comment saying something
like "Neil this is test #5, did it reach you?  It is a not
a real comment."






More information about the SeqFan mailing list