Doctor Fun updates mailing list information

Logging in, subscribing or unsubscribing: do all of your stuff at

Find out the what, how, who hosts, who reads and where previously for this list here. You can also find out what other unrelated stuff resides here.

What is this?

A spin-off of my wish to spend more time studying and less time browsing a site I check out regularly (and hopefully explanations like this page).

How does it work?

The short answer

A cron job runs at least three hourly from midnight (and hourly during the day, from nine o’clock to five o’clock) on server time, which I believe is Eastern Time in North America. The times in the subject headers, however, are in UTC.

It's fairly simple, but saves me (and if you want, yourself) browsing the Doctor Fun pages so often.

The list of addresses it monitors is actually limited to a few pages. You will find the current list appended to the end of each posting to the mailing list. This is to prevent you receiving messages every day when a new cartoon goes up, and to make list keeping much simpler. For that, you want to sign up for Nicholas Barnard’s latest Doctor Fun cartoons mailing list as well (or instead).

The long answer

This new incarnation of the Doctor Fun website changes mailing list utilises the kind hosting services of ibiblio. Specifically, it uses the following.

  1. a shell script, a login to run it on, and that login’s cron table
  2. lynx -nolist
  3. diff -b -B
  4. mail

Habeas headers on the individual licence are no longer used because the mail version used doesn’t like -a nor -e switches.

This stuff is then all sent to the Mailman list, which has also been moved to the same server. Being on the same server as Doctor Fun saves some of the issues I was having on my previous setup.

From the above commands listed, some of you may have noticed that the way I use lynx doesn't supply any indications of links on the pages being monitored. This is by design; you'll have to go to the pages themselves to confirm if there are any links.

Where is this list hosted?

This page is hosted by ibiblio.

Thanks, guys!

Actually, on the subject of hosting, who reads this list, anyway?

This page has been read by Nikster.

Where was this list previously hosted?

This list was previously hosted elsewhere, at Below is the old stuff, for the curious.

Well, to make my posts to the Doctor Fun website changes list on my/this box, I use the following. (I didn’t want to clutter up the main listinfo page with the following.)

  1. a shell script (and my login to run it on!)
  2. lynx -nolist (the -nolist directive unfortunately gets rid of the URLs, but keeps the posting sizes much lower)
  3. diff -b -B
  4. mail
  5. my Habeas individual licence (the shell script adds the header when it invokes mail; surprisingly, the individual licence seems to not mind this use of the Habeas headers), to prevent false positives on spam filters
  6. cron

All this then posts to a non-commercial mailing list setup (with double opt in) using Mailman that I’ve got on the same box doing all this, making quite a nice supplement to my occasional visits to the Doctor Fun site. :)

There’s also spam prevention for the mailing list in that only my account on this box can post. Also, your address is, within reason, safe with me, unless of course a competant legal authority tells me to disclose it. :(

You’ll find a current list of watched Doctor Fun Page URLs appended to each message sent by the mailing list.


It’s come to my attention that there is one major limitation to the above setup. It hasn’t been addressed in full yet.

The first thing to keep in mind is that this list is basically automated, so expect strange posts to happen occasionally, especially as described below.

Basically, if this box has a temporary outage in its Internet access, lynx -nolist -dump will give a blank output. That may mean massive diffs if the access goes down just before the page is checked, and the first successful lynx -nolist -dump run after the link goes up again. If you wanna give me a quick pointer to fix it, that would be great, thanks. :)

In the meantime, rsync -av is used daily at 0701 NZ time to reduce the effect of this problem. It doesn’t, however, solve the problem if the web server daemon goes down on localhost. Yes, I do need a more stable server, which is something I am working on.

Sorry for the trouble.

What other stuff is around here, by the way?

Well, there’s my page on Section 27(1) of the Copyright Act 1994 (NZ), including (as I write) a proof-of-concept digitisation of the (thus, apparently, I'm told, public domain because of the Act) Mazengarb Report of 1954.

Otherwise, sorry, I think that’s it for the time being. Enjoy.

© Copyright 2003, 2004 and 2005 by Jonathan Ah Kit ( This document’s XHTML 1.0 Strict-validated, and was last modified on Wednesday, 5 February, 2005.