Re: RC: Re: RE: Limestone Challenge; Announcement

[Ben Turner <bturner@ida.org>]

>At 04:22 PM 5/19/00 -0700, Duncan Fletcher wrote:
>>Go for it. In all the lists I am on, I have had the need for html only a
>>couple of times - that on another list for drawings related to gait
>>analysis. Of all the list serve mail I have received, only about twice has
>>anything useful been html or attachments (for some folks horse pics may be
>>useful, but I don't need to see a picture of somebody's cute new foal).

On Sat, 20 May 2000 22:54:21 -0700, David LeBlanc wrote:
>Stripping attachments isn't hard, but stripping HTML tags without causing
>what a friend refers to as 'random acts of terrorism' can be really
>difficult.  
-<snip>
>Plus, you're running the list out of a UNIX box (maybe Linux - can't tell
>without a good bit of poking), which means implementing anti-virus measures
>on the listserve itself will be difficult.  Additionally, a HTML
>interpreter will play all sorts of games with you depending on character
>sets, etc - you can't just do a:
>
>s/<[^>]+>//;
>(perl for get rid of anything that looks like an HTML tag), since %3a is
>the same as '<'

     Much of the HTML on the list is just duplicating the text body.
The text bodies are usually appropriately identified with MIME compliant
headers including "Content-Type: text/plain;" while the offending HTML
section headers usually include "Content-Type: text/html;"  Therefore,
much could be accomplished at the MIME section level without relying on 
detecting/stripping HTML within the text.  MIME aware PERL modules
and mail utilities exist that might streamline this process. 

     I support the notion of dumping all attachments _and_ HTML.

Best,
Ben Turner
bturner@ida.org