Albert Meyer albert at
Sat May 26 04:25:07 UTC 2001

At 06:18 PM 5/25/01 -0400, Steve Sobol wrote:
>Apples != oranges. You can avoid reading the binaries groups, and server
>can refuse to carry them.
>Once an e-mail is dumped in your mailbox, you can't automagically delete
>it if
>you are Joe Sixpack Average NetUser. You either have to go in with your
>client and download and delete it, or ask your ISP to nuke the message.
>*My* experience is that people who weren't expecting the big files from
>the yahoos that send it to them think their e-mail is broken until they
>and I check their mail spool and see there's a big file (obviously the
>is slightly different if they were expecting the message, but there are
>a lot of
>times when I hear that someone's friend or relative just dumped a big
>mail in
>their box). Large files are an annoyance to the customer in cases like

That's why the big-email dilemma is such a... well, such a dilemma. If you 
disallow huge emails they complain. If you allow them, they get stuck and 
need help. It was even worse when every 1user was using Outlook 97, which 
routinely choked on any email over 500K. We (my former employer) finally 
started charging $10 to clear out a stuck mailbox. After a time or two they 
would learn to use the web interface and clear it out themselves. A 
technical solution (like the one mentioned earlier) would be great, but it 
would need to work both ways, otherwise it wouldn't help until many people 
implemented it. Converting outgoing attachments to a link would be great, 
but it would also need to convert incoming attachments to a link. Incoming 
is the worse problem IMO. I'll add that to the list of things I plan to 
work on if I find myself without 60 hours/week of work that I get paid for.

More information about the NANOG mailing list