On    Sat,    9    Dec     1995    10:52:17    -0500    Mark    Hunnibell
<[log in to unmask]> said:
 
>I am not sure if I understand  you here. It sounds like you are agreeing
>that setting  subscriptions to FULLHDR just  to be able to  get the MIME
>header lines  is a solution that  creates a disk space  problem for most
>system administrators
 
No, I am saying that I disagree completely.
 
>Well...  forcing  everyone   on  the  planet  to   change  their  system
>configuration to allow more than 17 hops in a header seems like a rather
>excessive position
 
No, it's a perfectly  natural position, since a hop count  of 17 makes no
sense in  1995, LISTSERV or  no LISTSERV. There  has been the  exact same
problem with  TCP/IP TTL values. Sites  with ancient TTL limits  were cut
off as the  Internet expanded. Many of them whined  that the solution was
to expand the Internet in a different  way so that they would not have to
fix  their software.  They  also  predicted that  if  people didn't  stop
expanding the  Internet this  way, the  new sites would  just be  cut off
because  people simply  wouldn't  increase their  hop  counts, and  other
providers would  come along which  *would* be  happy to address  the very
important issue of the hop count.  The Internet kept expanding the way it
needed to, and people increased their hop counts.
 
>Lets  face it  too that,  as  good as  LISTSERV  is, if  you ignore  the
>real-world requirements in favor of adhering to an abstract philosophy,
 
Every  new  option increases  the  complexity  of  the software  and  the
confusion level of the users. Every day hundreds of people will ask their
helpdesk what  is the  difference between MIMEHDR  and FULLHDR.  There is
also a resource cost for every new type of header. Most sites make no use
of SHORTHDR, and having both FULLHDR  and MIMEHDR would double the number
of files in their mail spool.
 
In real world terms, you are asking for a new option to bypass an ancient
default in ONE product, sendmail. Furthermore  that limit is only ever an
issue in  the rare case  of peered lists  where the recipient  is several
peers removed from  the posting site, and the default  has been increased
in recent versions. I  am telling you that this is  a problem in sendmail
that has been fixed in recent versions,  and that the solution is for the
sites with  the ancient  default to  update either  their config  file or
their sendmail. If they aren't willing  to do either, there will be other
providers that *will*  be willing to invest an entire  5 minutes updating
one line  in their  config file so  that their users  can have  access to
remote LISTSERV lists.
 
  Eric