home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: alt.hackers
- Path: sparky!uunet!caen!sol.ctr.columbia.edu!startide.ctr.columbia.edu!seth
- From: seth@startide.ctr.columbia.edu (Seth Robertson)
- Subject: Re: Hack Question...
- Organization: Columbia University Center for Telecommunications Research
- References: <1992Sep10.213216.23658@ctr.columbia.edu>
- Message-ID: <1992Sep11.081606.5286@ctr.columbia.edu>
- Sender: news@ctr.columbia.edu (The Daily Lose)
- Approved: <god@ctr.columbia.edu>
- Date: Fri, 11 Sep 1992 08:16:06 GMT
- X-Posted-From: startide.ctr.columbia.edu
- X-Posted-Through: sol.ctr.columbia.edu
- Lines: 99
-
- In article <1992Sep10.213216.23658@ctr.columbia.edu> foo@bar.edu writes:
-
- >Anyone know how to get rid of the little annoying "X-Posted-From:"
- >line that sol.ctr.columbia.edu likes to append to it's posts...It's
- >REALLY annoying.
-
- Speaking as the person who implemented those restrictions, I know of
- no way to defeat those header lines for articles POSTed via NNTP. The
- system name is determined from the TCP/IP connection via
- getsockname(2) and is *APPENDED* to the headers in anne.jones via
- defhdrs.awk. It is the appending this which makes it hard to defeat
- since if you try to override the headers like you normally do for all
- other headers, they would just appear twice, making it utterly obvious
- to all and asundry that you are an idiot.
-
- The backup to this tracing is a mostly duplicate article which is
- posted locally. Yes, I DO make copies of all articles which are
- posted. I like to maintain an idea of what my site is being used for,
- but I have never used, and hopefully will never be forced to use, this
- information to restrict sites.
-
- Now if you want to know *why* I did this, it is for two reasons. The
- first, and by far the most important, is that I got fed up with people
- forging articles from my site; articles which were so annoying to
- other people that I was asked to trace them, articles which were
- boasts (to this very newsgroup, if I am not mistaken) saying that it
- was impossible for the articles to be traced (it is this last which
- annoyed me the most since a lot of even stupider idiots fingered sol
- and since I was logged on, assumed that I posted the article and
- bombarded me with mail). The other reason is somewhat less important:
- load. Sol is being greatly overloaded with people reading and posting
- news articles (401 sites reading 119630 articles) and I will be ****ed
- if I am going to add to this load by making it easy for people to
- forge articles from my site.
-
- This is actually curiously anomalous since I support the idea of people
- being able to post anonymously. There are certainly good reasons why
- someone might want to do this, but unfortuantly the idiot factor
- overwhelmed me. In any case, there are still plenty of easy ways to
- post anonymous without using sol (and no, I will not tell you unless
- you are from my site, ask your own sysadmin.)
-
- If any of you are still with me, I would like to take this time to
- urge you to lay off sol. The remote news reader loads imposed on it
- recently, due an unfortunate posting, will probably force me to (very
- reluctantly) impose maximum usage limits on sol.
-
- ObHack1: The hack above--contact me if you are a news admin running Cnews
- and would like my changes.
-
- ObHack2: My two favorite subroutines from Xkernel 2.0 (yes, I am
- working on it; expect to see it sometime after the Epoch :-)
- They are small hacks, but ``No hack is too small'' :-)
-
- print_pattern ()
- {
- [...]
-
- echo "Possible choices include:"
- for f in $1*
- do
- c=`echo "$f" | sed s/$1//`
- echo -n "$c "
- done
- echo
- echo
- }
-
-
- check_pattern ()
- {
- [...]
-
- ok=0
- for f in $1*
- do
- if [ "$1$2" = "$f" ]
- then
- ok=1
- break
- fi
- done
-
- if [ $ok -ne 1 ]
- then
- cat << EOF
- WARNING! This choice is not included in this distribution!
- If you answered correctly, you will have to generate the
- information yourself.
-
- EOF
- ask_to_die
- fi
- }
-
-
-
- -Seth Robertson
- seth@ctr.columbia.edu
-