From: IN%"email@example.com" "Chuq Von Rospach" 2-FEB-1997 17:30:10.54
>I can't speak for AOL on this, but looking at the outside, I see a
>*lot* of address churn from AOL. Folks sign up for AOL, and within a
>few days/weeks/months, the account dies or I get mail saying "I have a
>real address now, please change my subscription...." (real is my
>editorial, but it's suprising how many are pretty close to that). There
>*is* a strong case of AOL being a place people start and outgrow, from
>what I see, or who show up, poke around, and go away again.
That's about what I suspected.
>There *is* a place that's worse than AOL on this show-up-and-disappear
>churn. It's worldnet.att.com. I don't get a *lot* of subscriptions from
>that place (which indicates it's not overly successful...), but I *can*
>guarantee that almost all of the worldnet addresses that do sign up
>don't seem to stay long. That indicates to me that service is not
>terribly well liked by users, because I see lots of running for the
That's interesting in terms of Brad's worries re: AT&T proving
net service for lots of people. As I mentioned, I doubt that they're
going to provide enough _unnecessary_ handholding to cause too much
of a problem... on the other hand, they also seem likely to be not
providing even the _necessary_ handholding stuff.
>> I would personally attribute this to A. failures in the educational
>Don't necessarily blame AOL for this (or *just* AOL) -- first thing ANY
>of my lsits do when someone signs up is send them instructions. If the
>first thing *they* do is throw those instructions away unread, no
>educational system AOL comes up with is going to make significant
>inroads, either. As I like to say, throw a horse in the middle of the
>river, and it'll still die of thirst if it refuses to to open it's damn
Whoops... I should have been clearer here. I was referring
to the _US_ educational system; it seems to me to encourage this
attitude, particularly in the public schools. (As I stated, I'm
not as familiar with non-US educational systems, but I suspect the
same is true in general.)
>And let's also keep one thing in perspective: in general, we're talking
>about the 1-2% of the list that causes problems. I'd say 90% compliance
>is an easy target here, and on most lists, it's probably over 95%. It's
>the very noisy few driving us crazy here, and some of them simply won't
>be reachable no matter how many batteries you attach to your cattle
Quite... and as lists get potentially larger as the Internet
expands, this means there's going to be more of a problem, let alone
the proportion-of-newbies problem I also mentioned.
>We also need to keep in mind that we want to design systems that don't
>*screw over* taht 95% of non-trouble-makers in search of a magical
>solution for that last few percent who won't cooperate anyway... Let's
>not make it tough for people who do behave, or turn them into
>troublemakers -- shades of the good old days of software copy
>protection that assumed all of us were pirates and made it hard to
>actually USE the products legally... We need to keep the entire
>situation in view, lest the cure be worse than the disease...
Agreed. I'm not proposing making things _purposefully_
difficult in general, just that the large online services should
focus less on user-friendliness and more on functioning well in
other ways - including functioning better for users that know what
they're doing, or at least are willing to learn.
>Handholding, yes. Setting things up so people can fire up a drag racer
>before they have their drivers license -- those are the problems you
>need to watch out for in making things "easy", beacuse a few of those
>people are going to fire up that drag racer and proceed to roll it over
>into the grandstands. It's not *just* users taking themselves out, they
>tend to take others with them. (if they just were to drive that drag
>racer over a cliff, I'd call that "improving the gene pool".... grin)
Chuckle... you might find the DARWIN-L list (on yorku.ca) amusing. It's
largely stories of people doing exactly that - improving the gene pool
by taking themselves out of it.
>email mailbacks of subscription confirmations -- you don't get
>subscribed until I get mail back saying you really wanted to. The
>spammers have made that one necessary. It *also* gives me an
>opportunity to force documentation into their faces before they go on
>the lists, so there's no way they can claim they never saw anything. If
>nothing else, this just gives me a firmer stance on being able to
>enforce things, so there's less argument about people being kicked off.
IIRC, the latest version of Majordomo does this, with a
cookie. I'm not sure if it has that documentation included - that would
be a good feature idea.
>My lists have huge international audiences. In general, I have fewer
>problems overseas, but it's not absolute. They generally have a higher
>level of education, but language issues come into play, and frankly,
>twits will be twits, whether they're American, Swedish, Finnish or
>Italian. I get a few, but most of the problems I see in a place like
>Italy are more language issues where maybe easier/simpler documentation
>might help, while in Sweden where English is a priority second
>language, most of my problems tend to be the occasional twit.
Interesting. I wonder how much of it is that the higher level
of education (probably due to more monetary resources needed to get on
the 'net) and how much cultural differences resulting in less overt
cluelessness/twitdom and more "suffering" in silence?
>There are no easy answers. We've implemented all of those.... But that
>doesn't mean we've done all we can...
Good point. Noticeably, all the easy answers deal with things
directly under our control; the more difficult stuff is dealing with
the doings of others.