At 04:55 PM 3/19/2003 -0500, you wrote:
At 12:29 PM -0800 3/19/03, Steve Schear wrote:
At 01:13 PM 3/19/2003 -0500, you wrote:
Anybody who's tried to deploy software to end users (especially software
that involves retraining) knows 400M is the wrong place to attack. And
even a well-hidden Proof-Of-Work doesn't help here - your phone WILL ring
when people ask why it's taking 3 minutes to send mail.
Not if they don't have my phone number. Which they probably don't if
they haven't contacted me before.
Is that supposed to be a joke, or do you truly not care whether your
system is useable, so long as it has a nice architecture?
If my bounce mail reply (sent when an unknown sender contacts me) is
sufficiently simple for many/most users, who wish to take even a minimal
amount of extra time to contact me, works then that's good enough for
me. For others, who cannot or will not take the extra time/effort, I think
I can live without their communication.
I've never said I'm *STRICTLY* against a sender-side solution. My generic
criteria for a sender-side solution is: It has to be something that my
mother the Hotmail user can deploy without me or my brother having to
make a trip home to install for her. Windows 98 on a several-year-old PC.
This is not necessary for initial deployment. After all the web didn't
seem like it was ready for my grandmother 10 years ago.
Ten years ago the web was something entirely new--it wasn't replacing
anything, and it didn't have to interoperate with anything. The business
plan for creating something brand new is *way* different than the plan for
replacing something in use by hundreds of millions of people. People are
much more willing to try something new than they are to get rid of
something old--no matter how broken. To replace something you have to
show them very significant benefits. Your benefits depend on everyone
choosing your system over any other. When the number of people using a
system numbers in the thousands, that's not a big deal. When the number
of people using it numbers in the hundreds of millions, that's really bad odds.
Look at Kazaa. It takes extra effort to install and learn but 10s of
millions have done so because they see a perceived benefit. When Napster
first arrived most people who heard of it ignored it because they didn't
know or see see the value proposition. As time went by more people learned
why they wanted to spend the extra effort to download, install and learn to
use it.
These P2P systems didn't need millions to work but only 1000s or
10,000s. I'm OK, from my perspective with a little, initial, Balkenization
of email if it helps me. I don't care what it does or does not do for
others. It seems to me you're trying to come up with a solution that can
be plugged in without disadvantaging some users with minimal tech skills
(like our mothers). I'm saying I'm willing to do that if it helps ME.
steve
_______________________________________________
Asrg mailing list
Asrg(_at_)ietf(_dot_)org
https://www1.ietf.org/mailman/listinfo/asrg